Arxiv:2103.07712V2 [Quant-Ph] 27 Apr 2021 Same Time, Einstein Podolsky and Rosen Pointed Tum Mechanics of Open Systems [9] (See Also [10])

Ryszard Horodecki∗

International Centre for Theory of Quantum Technologies, University of Gdańsk, Wita Stwosza 63, 80-308 Gdańsk, Poland and Institute of Theoretical Physics and Astrophysics, National Quantum Information Centre Faculty of Mathematics, Physics and Informatics, University of Gdańsk, Wita Stwosza 57,80-308 Gdańsk, Poland

Dedicated to memory of Roman Stanisław Ingarden on his centennial birthday “. . . the quantum information theory is not only scientiﬁcally interesting subject, but is a practical need”

R.S. Ingarden

This article reviews the extraordinary features of quantum information predicted by the quantum formalism, which, combined with the development of modern quantum technologies, have opened new horizons in quantum physics that can potentially aﬀect various areas of our live, leading to new technologies such as quantum cybersecurity, quantum communication, quantum metrology, and quantum computation. topics: quantum cryptography, quantum entanglement, nonlocality, entanglement witness

1 Introduction application of quantum information to unforge- able quantum money. Unfortunately, both dis- The concept of quantum information was born on coveries were ahead of their time and passed un- the border between quantum mechanics and in- noticed. Three years later Holevo proved [6] that formation theory science. The stunning success of there is a bound for our ability to access classi- the former has led to think that the concept of cal information from quantum systems which con- information cannot be separated from the math- firmed earlier Gordon’s [7] and Levitin’s [8] con- ematical structure of quantum formalism that jectures. This strengthened the conviction that imposes fundamental constraints on the form of Shannon’s communication theory is incomplete, in physical laws. a sense that it did not consider the transmission Already in the 1930s, von Neumann defined of all physical information carriers such as quan- entropy [1] for quantum states as an analogue tum particles. A few years later, Ingarden, a Pol- of the classical Boltzmann-Gibbs entropy, which ish mathematical-physicist, published a work enti- later turned out to be the quantum counterpart tled: “Quantum information theory” in which of Shannon entropy [2] – the concept underlying he proposed a quantum generalization of Shan- of classical communication theory. At about the non’s theory in terms of the generalized quan- arXiv:2103.07712v2 [quant-ph] 27 Apr 2021 same time, Einstein Podolsky and Rosen pointed tum mechanics of open systems [9] (see also [10]). out the unusual features of quantum formalism However, it was only a series of seminal papers that seemed to lead to the conclusion that quan- [11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, tum mechanics is incomplete [3]. In 1970, two 24, 25, 26, 27] that revealed specific features the young physicists, Park [4] from the Department of quantum code of nature pointing to the quantum Physics at Washington State University and Wies- origins of information. ner [5] from Columbia University in New York, independently analyzed the physical implications There were various reasons for the relatively of quantum formalism. While the former discov- late advent of the quantum information era ered a fundamental limitation on copying quan- crowned with the bilding of Shannon’s quantum tum information, the latter discovered the first theory (see [28]). In particular, the unusual suc- ∗e-mail: [email protected]

1 cess of Shannon’s theory led to the belief that the momentum operators. The structures and mu- laws of physics could be derived from information tual interrelations of noncommutative observables processing as a purely mathematical concept de- bring deep questions concerning the properties of tached from physical information carries. On the the quantum systems related to the fundamen- other hand the identification of peculiar features tal principles: uncertainty and complementarity. of quantum information such as monogamy of en- The first one limits the precision of the statis- tanglement [29, 30, 31] required advanced quan- tics of the results of two complementary observ- tum technologies. Additionally, the obstacle was ables, such as position and momentum [34]. The the abstract, mathematical and non-intuitive na- complementarity principle says that two quantum ture of the standard quantum formalism, which observables cannot be measured simultaneously, looked like inscription, not all predictions of which and thus provide “independent” information about were entirely clear even for its fathers. physical systems [35]. Contrary to classical theories, quantum mea- 2 Quantum inscription as a paradigm for surement is active. It creates properties, does it quantum information randomly, and can change state if the latter is not specially tailored for a given measurement. The measurement does not always provide information Roughly speaking, quantum inscription is an in- about state but it can be part of a quantum oper- struction – a set of prescriptions that determine ation. Any state % defines the probability distri- the way of probabilistic prediction of the results bution as the mapping assigning to each measure- of future measurements in laboratories [32, 33]. ment result i the probability pi of that measure- Each physical system corresponds to complex ment result (the Born rule): vector space Hilbert H equipped with the linear scalar product h.|.i such that the space is complete pi = Tr[Πi%] (2) with respect to the norm P where {Πi}, Πi = I are elements of a positive kψk = phψ|ψi. i (1) operator-value measure (POVM) and I is unit operator. In particular, if Πi is projector operator H S n The space of system compound of sub- then the generalized measurement correspond to S ,S ,...,S H = systems 1 2 n is a tensor product the von Neumann measurement, which completely H ⊗ H ,... ⊗ H 1 2 n of the Hilbert space of sub- determines the post-measurement state. After the systems. The subsystems can represent distin- measurement with the outcome i, the system goes guishable particles, various complex objects, e.g. to the post-measurement state atoms, molecules, or different degrees of freedom of the same object, e.g. photon polarization and 0 −1 %i = pi Λi(%) (3) propagation modes. The central object is the wave function (state where Λi(%) = Πi%Πi is particular positive su- vector) |ψi with the unit norm kψk = 1, which is peroperator which clearly maps positive operators 0 an element of a Hilbert space. It contains all prob- to positive operators and normalization of %i re- abilistic information about the system and sat- quires the condition to be met Tr[%Πi] = Tr[Λi(%)] ∂|ψi 0 isfies the Schrödinger equation: i} ∂t = H|ψi, where Λi(%) = pi%i. The most general phys- where H is linear self-adjoint operator called ically implementable map is a completely posi- Hamiltonian. The symbol % denotes the state tive map Λ which satisfies condition: Λ ⊗ In ∈ n n of the system about which we only have par- B(H1 ⊗ C ,H2 ⊗ C ), where B is space of pos- n tial information. It can be described by a Her- itive maps between the Hilbert spaces H1 ⊗ C n mitian positive semidefinite operator with unit and H2 ⊗ C , In is unit operator on n dimen- trace: % = %†, % ≥ 0, Tr(%) = 1 where trace sional Hilbert space Cn. If in addition Λ is trace- P Tr(%) = khφk|%|φki and sum runs over diago- preserving it determines quantum channel which nal elements in arbitrary orthonormal basis {φk}. play a central role in the processing of quantum The symbol U stands for unitary operations that information [28]. Any completely positive map on transform states, and in the case of pure states, a system S in a given state % can be realised via they keep the scalar product preserved. unitary interaction of S with some other system Observable quantities correspond to Hermi- (ancilla) in a pure state followed by von Neumann tian linear operators O acting on the state space measurement and final partial trace. This fact H. In contrast to classical observables the quan- comes from so called Stinespring dilation theorem tum ones can be noncommutative: [O1,O2] = [32]. O1O2 − O2O1 6= 0. The most familiar example The crucial difference between the quantum is [Q, P ] = i where Q and P are the position and description of physical reality and the classical one

2 is the principle of superposition: if |Ψ1i, |Ψ2i are 3 Quantum bit – the unit of quantum in- system states then their superposition; formation

The concept of qubit appeared for the ﬁrst time in |Ψi = a|Ψ i + b|Ψ i 1 2 (4) the context of the theory of quantum information transmission [23] as a two-level system, the state is also in good state, provided that a and b are of which can be written as a superposition of two chosen so that |Ψi is normalized. base states |0i and |1i

The prediction power of quantum inscription is |Ψi = a|0i + b|1i (5) astonishing: “All our experience so far using quantum theory seems to say: What is pre- where a and b are complex numbers, |Ψi ∈ C2 dicted by quantum formalism must come (two-dimensional Hilbert space). to the laboratory” [36]. In the early 1970s, Contrary to the classical bit, the qubit repre- it seemed that all possible predictions of quan- sents a continuum of possible states defined by tum inscription had already been recognized. The its wave function, which can be visualized by papers of Einstein, Podolsky and Rosen [3] and the two-dimensional Bloch sphere with two real Schrödinger [37] were initially treated rather as a parameters θ and ϕ where a = cos(θ/2), b = mathematical artefact detached from its physical sin(θ/2 exp(iϕ)) where 0 ≤ θ ≤ π, 0 ≤ ϕ ≤ 2π. implications. Ironically, it was them who drew at- For illustration, consider a photon as a paradig- tention to the extraordinary implications of quan- matic example of a qubit. It requires a Hilbert tum inscription, which revealed the existence at space H which is a tensor product H = Hprop ⊗ a fundamental level of a subtle order governed Hpol, where Hprop represents the photon propaga- 2 by quantum information. In the classical world, tion modes while Hpol = C describes the photon quantum information is “unspeakable”. It cannot polarization modes. If one disregards the prop- be written with discrete symbols, e.g. on a tape of agation modes, the photon can be treated as a a Turing machine. So far, there is no commonly photonic qubit in polarization degree of freedom. accepted definition of quantum information. Consider now a photon in the superposition For our purposes, it is convenient to adopt of the base states |0i ≡ | li, |1i ≡ | ↔i corre- the following interpretation: Quantum infor- sponding to vertical and horizontal polarization mation is what is carried by quantum par- |Ψi = sin Θ |0i+cos Θ |1i. If we direct it to a verti- ticles and the wave function ψ is its math- cal polarizer, it will change to one of the states |0i 2 ematical image [38]. or |1i with probabilities p0 = T r[Πo%] = sin Θ, 2 p1 = T r[Π1%] = cos Θ respectively, where Π0 = Quantum information (QI) can be processed |0i, h0|, Π1 = |1i, h1| are projectors and density (manipulated) [32, 39], using combinations of uni- matrix of the state |Ψi is given by tary operations and measurements. QI is the source of quantum resources [40] such as entan- sin2 θ sin θ cos θ % = |ΨihΨ| = glement [36, 41, 42], steering [43], quantum corre- sin θ cos θ cos2 θ (6) lation beyond entanglement [44], quantum coherence [45], asymmetry [46]. It allows to perform where the diagonal elements are interpreted as nonclassical tasks such as quantum cryptography the probabilities of the basis state, while the off- [13, 16, 47, 48], teleportation [19, 25, 26, 27], quan- diagonal elements represent the coherence of the tum computing [15, 14, 18], not feasible with clas- basis states. sical resources. QI is resource for quantum metrol- If we now place a specially cut birefringent ogy [49], computational complexity [50, 51, 52]. crystal with the optical axis at an angle of 22.5 However, this subtle resource has a very un- degrees on the path of a vertically polarized pho- pleasant feature. As one knows, non-diagonal el- ton, the photon will be in a state of linear su- ements of the density matrix % called coherence perposition (Fig. 1). This is nothing but the in the state %, provide information about quan- photonic realization of the Hadamard H gate. It tum interference. Unfortunately, as a result of has no classical counterpart and plays a funda- the system’s interaction with the environment, the mental role in quantum information processing in- process of decoherence [53] occurs, which causes cluding quantum computing. Note that arbitrary disappearance of non-diagonal elements of density photonic wave plate operations for photonic polar- matrix of the state. Reversing the degradation of ization qubits realizing Hadamard, Pauli-X, and quantum information still remains a great chal- rotation gates were implemented on the chip [54]. lenge for effective processing of quantum informa- In Fig. 1, B1 and B2 denote the computation tion. base and the Hadamard base, respectively, which

3 22,5 1 H = 1 1 45 √2 [1 −1 ]

↔ 1 1 | 〉= |↔ 〉+ |↕〉

|↕ 〉 √2 √2 ↔ 1 1 |↕ 〉 H|↕〉= |〉 = |↔〉+ |↕ 〉 √2 √2

1 1 |↔ 〉 H |↔〉= | 〉↔ = |↔〉− |↕〉 √2 √2

B 1 B 2

Figure 1: The photonic realization of the Hadamard H gate. B1 and B2 denote the computation base and the Hadamard base. are mutually unbiased, i.e. they are mutually ex- chine would have to be nonunitary and non-linear, clusive. Perfect information about the polariza- which is forbidden by the linearity of quantum tion along the selected axis implies that there is formalism. Thus copying destroys the state and no information about the polarization along the it cannot be reconstructed from a single copy. axis rotated by 45◦. This is a purely quantum Hence the quantum signals cannot be noiselessly mechanical effect resulting from the fact that the amplified. Later the limitation for the unperfect vectors |0i, |1i and |+i ≡ | %.i, |−i ≡ | -&i are cloning in terms of the so called fidelity function the eigenstates of the Pauli operators σz and σx, f(%out) = hΨ|%out|Ψi measuring similarity of the respectively, which do not commute, ie. [σz, σx] = state of either of the two outcome registers has σzσx − σxσz 6= 0. been provided within the framework of imperfect There have been many proposals for the phys- quantum cloning machines [66, 67]. There is dual ical realization of a qubit on quantum dots [55] the non-deleting theorem, which states that, in electron spins [56], semiconductor spin [57], su- general, given two copies of some arbitrary quan- perconducting charge qubits based on Josephson tum junction [58, 59]. Remarkably it has been demon- state, it is impossible to delete one of the copies strated, that linear optics is sufficient for effi- [68]. In the above mentioned paper Holevo [6] cient quantum information processing with pho- proved an fundamental theorem that sets an up- tonic qubits in two optical modes (such as hor- per limit to the amount of information available izontal or vertical polarization) [60, 61]. There about a quantum state. It implies that with the has recently taken place a quite progress in paral- help of one qubit is impossible to send more than lelized quantum information processing which in- one bit of classical information. cludes tailored quantum memories to simultane- Quite unexpectedly, it turned out that there is ously handle multiple photons [62]. also a restriction on the possibility of generating of quantum superposition. Namely, it has been inde- 4 Fundamental limitations on quantum pendently shown [69, 70] that there is no universal information processing probabilistic quantum protocol generating superposition of the two unknown states. Interestingly, Already in 1961 Wigner pointed out that the ex- a probabilistic protocol generating a superposition istence of self-reproduction in the quantum world of two unknown states having a fixed overlap with is unlikely [63]. In 1970. Park [4], and later Woot- a known pure reference state has been proposed ers and Żurek [64] and Dieks [65] proved that it is [70]. This protocol has been carried out experi- impossible to build a quantum machine that can mentally in a three-quadrant NMR system as well perfectly copy arbitrary unknown quantum state as on unknown photonic quantum states [71, 72]. Ψ:

|Ψi|0i|Mi 6→ |Ψi|Ψi|Mψi (7) 5 Quantum cryptography based on no- cloning where |0i means a blank state, while |Mi, |Mψi are machine state before and after cloning re- Parallel to Park’s paper on non-cloning, Wiesner, spectively. The process realized by such a ma- based on principle of uncertainty introduced the

4 concept of conjugate coding to make up quantum and practice enabling quantum hacking, e.g. The money [4]. This idea paved the way for the quan- Bright illumination Attack, Photon number split- tum information encryption Bennett’s and Bras- ting [79]. Therefore, QKD implementations are sard’s protocol (BB84) [13]. It has the following still in the testing phase and these gaps are iden- main three steps: tified. Stronger versions of BB84 were developed, such as the BB84 decoy state and protocols resis- 1. Alice sends randomly polarized photons tant to photon number breaking attacks [47]. As through the quantum channel in the se- a result, QKD protocols become more and more lected computing bases {B1} |0i, |1i and secure. Hadamard {B2} |+i, |−i; saves bases and bits. 6 Quantum entanglement – the most nonclassical feature of quantum information 2. Bob measures photons in randomly selected bases B1 and B2, registers bases and bits. As we have seen, already at the level of simple systems, the properties of quantum information 3. Via the classic public (authenticated) chan- differ substantially from those of classical infor- nel, Alice and Bob transmit their choices mation that can be amplified and copied. Much bases. When their bases match, they retain earlier, in the 30s, EPR and Schrödinger revealed the appropriate bits. a peculiar feature of quantum information in complex quantum systems rooted in the principle Thus, they receive a raw key that requires fur- of superposition called entanglement. According ther processing. To check for eavesdropping, they to the quantum inscription, the state space HS calculate the quantum bit error of a randomly se- of the quantum system S compound from dis- lected data subset that they reveal each other via tinguishable subsystems S1,S2,...Sn is given by the public channel and check if the error (percent- HS1 ⊗ HS2,... ⊗ HSn which is the tensor product age of mismatched bits) is below a certain thresh- of the Hilbert space of the subsystems. old value. Using classic post-processing protocols We say that a pure state is entangled if it can- such as error correction and privacy amplification, not be written as the product of the states of the they generate the final secure key. individual subsystems Since 1992, when Bennett and Brassard and colleagues demonstrated the first 32cm quantum |Ψi12...n 6= |φi1 ⊗ |ψi2 ... ⊗ |χin (8) distribution of the key in free space [73], there has been tremendous progress in the development In general a mixed state % of n systems is en- of quantum cryptography in free space and in tangled if it cannot be written as a convex combi- fiber. There is a continuous improvement of cryp- nation of product states tographic keys over long distances [74] as well X i i % 6= %sep = pi%1 ⊗ · · · ⊗ %n (9) as an increase in key generation speed using sin- i gle photon detectors [75]. Quantum key distribution (QKD) networks were established in the In particular, for any two-part pure entangled US, Austria, Switzerland, China and Japan. and state |ψi12 ∈ H1 ⊗ H2 there exist orthonormal the European SECOQC network [76]. Due to ex- Schmidt bases {φii, {χii in H1, H2 respectively ponential signal attenuation and decoherence, the such that: effective distribution range of the quantum key d X of terrestrial networks is limited to 300 km [77]. |Ψi12 = ci|φii ⊗ |χii (10) In cosmic space, both of these factors are many i times weaker. In 2016, the first satellite distri- where the summation takes place on the smaller bution of the BB84 protocol was performed us- dimensions of the two systems d = min(d1, d2). ing a one-time key cipher via the Micius satellite In particular, the two-part maximally entangled at intercontinental distances, thanks to which the 2 state in the space H1 ⊗ H2 with the dimension d photos of Schrödinger and the philosopher Micius is defined as: were safely transferred between Vienna and Bei- d jing [78]. 1 X |Ψmaxi = √ |φii ⊗ |χii (11) Despite the enormous advances in quantum d i cryptography, there are still some problems related to the fact that practical implementations In particular there is a two qubit entangled of quantum key decomposition use realistic pho- state: tonic qubits and imperfect single photon detec- + 1 |Φ i = √ (|0i1|0i2 + |1i1|1i2) (12) tors. This creates gaps between QKD theory 2

5 where {|0i, |1i} is the computational basis for important result was in particular discussed in the qubits. Using von Neumann entropy as a mea- context of quantum computing on NMR which op- sure of entanglement for pure states, it is easy to erates on highly mixed, separable states [93]. check that the above state above contains one ebit The discovery of Einstein, Podolsky, and of entanglement, i.e. the maximum amount of en- Rosen that entangled states could show “ghostly” tanglement that a system with dimension d = 22 correlations independent of distance, until the ap- allows. In general, for a system consisting of n pearance of John Bell’s famous work, was not pairs of entangled qubits and a Hilbert space di- given much interest. On the one hand, they were mension, d = 2n contains n ebits of entanglement. considered more philosophical than physical, on Most of the pure state vectors in a pure state two- the other hand, it was believed that such correla- part Hilbert space are not maximally entangled. tions could be simulated classically. For systems divided into more than two parts, the Schmidt distribution in general does not ex- 7 Photons entangled in polarization ist. However, many of the important states in quantum information processing take the form of To illustrate this phenomenon, consider the prob- a multi-part Schmidt distribution. Among them, abilistic generation of photons entangled in polar- three-particle W√ and GHZ states: |ΨiGHZ = ization degrees of freedom using Type-II down- (|000i +√|111i)/ 2 [80], |ΨiW = (|001i + |010i + conversion [94]. In this process, a high-energy |100i)/ 3 [81], which represent two different photon in an optical nonlinear medium (BBO types of entanglement that cannot be transformed crystal) is converted into two lower-energy pho- into each other through local operations and clas- tons that are emitted along the surface of two anti- sical communications (LOCC). Interestingly ex- correlated intersecting cones with vertical and perimental W-to-GHZ state conversion were re- horizontal polarization (Fig. 2). In particular, the cently demonstrated [82, 83]. photons emitted along the intersections cannot be Let us emphasize that the above mathematical assigned a specific polarization because we do not description of quantum entanglement between the know which cone they come from. We write it various degrees of freedom of complex systems is down as a quantum alternative adequate in a scenario where each subsystem (e.g. + 1 |Ψ iAB = √ (|0iA|1iB + |1iA|0iB), (13) qubit) can be individually addressed / manipu- 2 lated. In situation when one consider indistinguishable systems in connection with symmetri- where |0i, |1i correspond to vertical, and horizon- sation postulate the complete characterization of tal polarisation respectively. Here |Ψ+i is one of entanglement is still challenge. Many different ap- four canonical Bell-states (Bell basis) [95]: |Ψ±i = √1 (|0i |1i ± |1i |0i ) |Φ±i = √1 (|0i |0i ± proaches have been proposed with different entan- 2 A B A B , 2 A B glement definitions. Recently, Benatti et al. [84] |1iA|1iB), Now, if we direct the photons from the made an extensive comparative analysis of differ- entangled pair (EPR pair) in the polarization to ent approaches to the definition of entanglement the distant Alice’s and Bob’s laboratories, respec- of quantum systems composed of indistinguish- tively, who independently measure the polariza- able particles based on natural physical require- tion of the same type, it turns out that they get ments. anti-correlations 0 − 1 or 1 − 0. What is striking There are many ways to generate quantum is the fact that individual photons do not carry entanglement. Entangled states are most often any bit because their polarization is completely generated in the spontaneous parametric down- random [16], so local measurement results turn conversion and spontaneous four-wave mixing [85, out to be completely random too. EPR reasoned 86, 87]. It is intriguing that it is possible to en- as follows: If it is possible to “remotely” predict tangle together particles from two independent some property of a particle without interacting sources that did not interact with each other in with it, then this property must have existed be- the past [88, 20]. Another peculiar behaviour of fore, ie. before the measurement. They called entanglement called a sudden entanglement death it the “reality elements”, and from there they con- was described in a dynamic scenario. Namely, cluded that quantum inscription offered an incom- when two entangled qubits interact with natu- plete description of physical reality. ral reservoirs, the entanglement can disappear in a finite time while the coherence disappears 8 Nonlocality of quantum correlations. asymptotically [89, 90, 91]. The source of this Bell tests phenomenon is due to the fact that in finite- dimensional systems the set of separable (non- It was a serious objection that no one, includ- entangled) states has a finite volume [92]. This ing Bohr himself, was able to convincingly re-

6 extraordinary vertical cone

UV pump

BBO crystal

ordinary horizontal cone

Figure 2: Generation of photons entangled in polarization using Type-II conversion in the Bell state + |Ψ iAB fute. The Gordian knot was cut by John Bell Quantum mechanics predicts that the mean [11]. Namely, he formalized the concepts of reality value of the B observable√ satisfies the inequal- elements by introducing a model of local hidden ity |hBQM i| ≤ 2 2 which means√ that it breaks variables based on the following assumptions: i) Bell-CHSH inequality, where 2 2 is the so-called the measurement results are determined by the Tsirelson’s bound [97]. properties of the particle carried before and inde- The verification of Bell’s inequality based on pendent of the measurement, ii) the results deter- the assumptions of local realism proved to be a mined in one place are independent of any actions great challenge for experimentalists, as it required in the space-like separation, iii) the settings of lo- the closure of three loopholes: i) Locality demands cal apparatus do not depend on hidden variables that no signal traveling at the speed of light can that determine the results of local measurements. inform the distant detector of its settings or the These assumptions, as Bell showed, impose con- result of a measurement on the local detector be- straints on correlations called Bell’s inequalities. fore Alice and Bob complete the measurements; The key point is that they can be verified in the ii) Fair-sampling (or detector efficiency) demands laboratory regardless of any theory. that the sample of entangled pairs be a faithful Let us briefly illustrate the Bell inequalities representation of the entire ensemble being broad- by exemplifying the correlation of polarized en- cast; iii) Freedom of choice requires that the hy- tangled photons that were sent to the distant Al- pothetical local variable should not influence the ice and Bob laboratories along the z axis. The local choices of measurement setups on the part partners measure dichotomous observables ie po- of Alice and Bob. larizations that have only two values +1 or −1. The first ground breaking experiment that Each partner measures two such observables. Al- convincingly demonstrated breaking the Bell ice chooses the settings of detectors a, a0, Bob b, CHSH inequality and good agreement with the b0, which are unit vectors showing different angles predictions of quantum mechanics was performed in the x-y plane along which they can orient polar- by Aspect et al. [12]. In their experiment, en- izing filters. For each pair of settings, correlation tangled photon pairs were emitted by the process functions can be constructed: ha, bi, ha, b0i, ha0, bi of atomic calcium cascades. For the first time, ha0, b0i, where hi means the average of the product the authors used acousto-optical switches, which of outputs. On this basis, it is possible to build a pseudo-randomly changed the orientation of the new Bell observable B = a, b + a, b0 + a0, b + a0, b0. analyzers in a short time compared to the photon Now, if we accept the assumptions of local real- transit time and detection. They achieved more ism, in particular that each photon had a certain than 95% of the detection efficiency. polarization value (+1 or −1) before the measure- Only in 2015, a series of Bell tests based ment, it is easy to check that the absolute value of on quantum random number generators was per- Bell observable cannot exceed 2. Hence, we obtain formed, which closed both locality and fair- Bell-CHSH inequality [96]: sampling loophole in the same experiments [98]. Recently, two cosmic Bell tests with photons en- 0 0 0 0 |hab + a b + ab − a b ikl| ≡ |hBi| ≤ 2 (14) tangled in polarization were performed, in which

7 measurement settings were determined by real- and local partial traces the state is in the unique time photon wavelength measurements from high qubit form (13). This is an essence of the device redshift quasars, light emitted billions of years independent variant of the Ekert’s entanglement- ago; Thus, the authors closed two loopholes at based encryption protocol (E91) [16] (see Sec. 11). once: locality and freedom of choice [99, 100]. Quantum self-testing is a cornerstone of device in- However, these experiments failed to close the dependent quantum cryptography which is based fair-sampling loophole. Quite recently Pan et al. on the idea that only the output statistics of the [101] performed an impressive local realism test devices are enough to guarantee cryptographic se- that closes both locality and fair-sampling loop- curity without need of knowing the physical struc- hole and rules out common cause 11.5 years before ture of the devices (for example see [116]. Fi- the experiment, which largely closes the freedom nally there is a weaker variant of Bell inequalities of choice loophole. on composite systems that is still much stronger The interpretation of violating Bell’s inequal- than contextuality. This is based on the so called ity is still the subject of the discussions [102, 103]. quantum steering [43] in which we assume that for The Bell tests show that the quantum correla- one of the particles the dimension of the Hilbert tions cannot be explained using any theoretical space is known (much like in contextuality tests) model based solely on local variables. This par- while in the other is not. This leads to the so ticular feature of quantum information, which called semi-device independent quantum cryptog- has become known as quantum nonlocality (Bell raphy (see [117] and reference therein), [118]. nonlocality), provides the resource for device- independent quantum key distribution [104, 105, 10 Nonlocality and the principle of infor- 106] (see however [107]). mational causality

The discovery of quantum nonlocality shook our 9 Weaker forms of breaking realism perception of the foundation of quantum physics. Hence the natural question arose: Is there a non- While I’m not going to do a detailed review of the locality stronger than that predicted by quantum vast field of difference in Bell’s inequality, let me formalism? Is this the only description that al- mention two important related concepts. First, it lows for nonlocal phenomena consistent with spe- should be mentioned that violation of local realism cial relativity? In the 1994 paper, Popescu and by composed quantum systems has it’s a weaker Rohrlich (PR) [119, 120] took nonlocality as the quantum analog called quantum contextuality, ob- basic axiom and have proposed a model indepen- served with help of random measurements of spe- dent approach, consistent with special relativity, cially designed sets of quantum measurements pi- based on the conception of input-output black- oneered by [108] which has many further devel- box devices. In the approach the experiments of opments (see [109, 110, 111]) can be mathemat- Alice and Bob are space-like separated and each ically quantified [112]. Quite remarkably it have experiment is treated as a black-box. Then all the the so called state variant fully analogous to Bell physical information obtained in the experiment is inequalities [113] as well as state-independent one encapsulated in the joint probability P (a, b|x, y) which is valid for any state, and basically reports that Alice obtains a and Bob b when Alice inputs the nonclassicality of the sent of measurement in- x and Bob inputs y respectively. In the simplest volved [114]. case where x, y, a, b have only two possible val- The fundamental difference is that roughly ues, they must satisfy the constraints: a ⊕ b = xy speaking quantum contextuality can contradict where ⊕ denotes addition modulo 2. It is not classical realism only under assumption of some difficult to verify that PR nonlocality leads to bound on dimension of Hilbert space, while vio- algebraic breaking of CHSH inequality equal to lation of Bell inequalities via quantum states is √ 4 which drastically breaks Tsirelson’s limit 2 2. the phenomenon that is independent on that as- Does nature allow information to be processed us- sumption in general. This is why violation of the ing such super-quantum correlations? Remark- inequalities in many cases leads to the powerful ably the physical principle of information causal- concept of quantum self-testing [115]. In the case ity was proposed [121], which excludes such pos- of the inequality (14) self-testing means that in- sibility. The information causality principle can dependently of complexity of local systems (for be formulated briefly as: The message cannot instance one may assume that each of the observ- contain access to more information than ables in (14) may concern not polarisation but the amount contained in it. Contrary to its la- some other or even all of the photon internal de- conic form, this principle has strong implications: grees of√ freedom) the saturation of the quantum bound 2 2 guarantees that up to local isometries • It strictly determines the maximum value of

8 √ quantum correlations ≤ 2 2 of entanglement which express the fact that entanglement represents correlations that cannot be • it is fulfilled by both classical theories and shared by third parties [29, 30, 31]. This pecu- quantum mechanics liar entanglement trait not only provides the security of entanglement-based cryptography, but • it excludes the physicality of the super sheds new light on physical phenomena in many strong Popescu-Rohrlich correlations correlated systems [132]. It is significant that although the properties of Experiment implementations of the E91 proto- quantum and classical information are basically col have been made at ground stations [133, 134]. different, they both follow the principle of infor- Recently, both production and analysis of entan- mational causality. It should be noted here that gled states have been tested with the SpooQy nonlocal PR boxes although nonphysical provide satellite, which is a step towards the realiza- a conceptual tool in the modeling of nonlocality in tion of a cryptographic key generator based on the quantum physics and beyond [122, 123, 124]. entanglement in cosmic space [135]. Quite re- It is remarkable that the PR correlations are un- cently, the quantum key distribution has been an- der some circumstances much more powerful re- alyzed with a small block length, which is crucial source than quantum entanglement as they lead in entanglement-based quantum communication to trivialising quantum communication complex- [136]. It should be emphasized that the original ity [125, 126]. However they are weaker in another E91 protocol was prophetic as it suggested device- sense since in their language there is no room for independent cryptography [137, 105], based on nontrivial dynamics and continuous chance of set- Bell inequality breaking, which ensures that the tings of the measurements. data produced by quantum devices has a certain Finally it is worth noting that in the case of degree of secrecy, no matter how exactly the data three parties the concept of relativistic causal- was generated. ity that goes beyond the no-signaling paradigm is possible when space-time variables are explicitly 12 Canonical effects based on quantum involved [127, 128]. Quite recently the general ax- entanglement iomatic approach to causality of the evolution of the spatial statistic detection has been initiated Ekert’s work was important for another reason, [129, 130]. namely, it was the first to show that “ghostly” EPR correlations can be harnessed into something 11 Entanglement-based cryptography useful. Since then, entanglement has been viewed not as a curiosity, but as a real physical resource As mentioned above, quantum correlations, apart that can offer completely new unexpected effects. from nonlocality, have another feature – they are The breakthrough was the discovery of dual ef- random. It was intriguing that this random- fects, i.e. dense coding and quantum teleportation ness ensures the peaceful coexistence of quan- in which the ebit plays a central role, i.e. a pair of tum inscription predictions and special relativity, qubits in a maximally entangled state, distributed as partners cannot use the correlation to the in- between the sender and receiver. Remarkably stant telegraph. This specific “telegraphic no-go” both entanglement-based effects circumvent the has not yet had clear theoretical foundations, al- non-cloning and Holevo theorem. though recently an attempt to explain this phenomenon has been made [131]. 12.1 Super dense-coding As we saw, singlet-state photon pairs entangled generate anti-correlated random numbers at Suppose Bob wants to send to Alice two bits of distant locations. Ekert first noticed that the ran- information, using only one noiseless qubit. Ac- domness of these correlations could be used to cording to Holevo’s theorem, only one bit can be generate a secure cryptographic key and proposed transferred with one qubit. So Bob would need 1 the protocol E91 [16] based on the entangled spin 2 two qubits for this. Bennett and Wiesner showed particles in singled state and Bell’s theorem and [17] that if Alice and Bob have one ebits then it proposed implementation using nonlocal correla- is enough to send only one qubit to transmit one tions between maximally entangled photon-pairs. of the four messages (00,01,10,11) to Alice. To do Soon after, the Bennett, Brassard and Mermin this, Bob encodes messages using local different proposed a simplified protocol based on entangle- unitary operations U00, U01, U10, U11 on his qubit, ment without Bell’s theorem, and showed that it generating orthogonal Bell states (Bell base), and is equivalent to BB84. The security of E91 is due sends the qubit to Alice, which measures the com- to the fundamental property called monogamy bined two qubits. The four orthogonal Bell states

9 represent the four distinguishable messages. The including to continuum variables [142, 143]. first implementation of a super-dense photon en- Quantum teleportation, was demonstrated in pio- coding protocol was made by Mattle et al. [138] neering experiments by the Zeilinger [25] and De in which Bob performed unitary operations using Martini [26] teams. Furusawa and co-workers [27] a combination of half and quarter revolutions of independently carried out a unconditional telepor- the wavelet. The dense coding protocol was later tation on continuous variables (see in this context implemented in particular on atoms [139] and nu- [142, 143, 144, 145, 146]). Later, quantum telepor- clear magnetic resonance [140]. tation was demonstrated in many beautiful experiments [147, 148, 149, 150, 151, 152]. In 2017, a photon was teleported from Ngari ground station 12.2 Quantum teleportation to the Micius satellite (with an orbit from 500 to The most astonishing prediction of quantum in- 1400) [78, 153]. scription is quantum teleportation – a dual effect Quantum teleportation has been continuously to dense coding that demonstrates the remarkable researched for more than 20 years (see ref. [154]) power “exotic” combination quantum and classical due to its central role in the development of resources (see the fascinating story of the discov- quantum information processing including quan- ery [141]). tum computing [147, 155], the quantum internet and its relationship to the foundations of physics. This time Alice wants to send one qubit to Various generalisations of the original protocol Bob in an unknown state, but not by physical have been proposed. In particular, the original qubit transfer, having two classic bits at her dis- protocol was generalized including general tele- posal. Obviously, quantum information cannot be portation channel [156], multiport teleportation transferred with classical bits. Let now consider [157, 158, 159], teleportation with multiple sender- the situation if we provide partners with 1 ebit receiver pairs [160], telecloning [161]. of entanglement. Now Alice can perform a measurement on her two particles, i.e. a qubit in an 12.3 Entanglement swapping unknown state φ and a particle from the entangled pair. It is not hard to see that this mea- The peculiarity of multi-particle entanglement is surement is identical to what Bob made in high- that one can entangle particles that have never density coding. Alice gets one of four possible interacted with each other in the past. That such 1 outcomes with a 4 probability: 00,01,10,11. Hav- an effect may take place was suggested by the first ing two bits at her disposal, Alice can send infor- Yurke and Stoler (1992b) [88]. This idea was im- mation via the classical channel to Bob which of plemented in the pioneering paper: “Event-ready- the results she received. Depending on the result, detectors” Bell experiment via entanglement swap- ∼ Bob uses one of the transformations: U00 = I, ping. In this scenario, arbitrarily distant partners ∼ ∼ ∼ U01 = σx, U10 = σy, U11 = σz where σx, σy, Alice and Cecilia and Bob and David share entan- σz, are standard Pauli operators. At this point, gled EPR pairs of photons coming from indepen- his particle from the entangled pair it will be in dent sources: state φ. Note that Alice’s measurement provides + 1 no state information (the bits are completely ran- |Φ iAC = √ (|00i + |11i), dom), but is part of a quantum operation. So the 2 transmission of the qubit had to take place im- + 1 |Φ iBD = √ (|00i + |11i) (15) mediately at the moment of Alice’s measurement. 2 There is no conflict with special relativity here be- The system is then described as cause quantum inscription predicts that any operation on one subsystem does not cause measur- + + |Φ iAC ⊗ |Φ iBD (16) able changes on the other subsystem regardless of the state of the entire system. Note that there Now Cecilia and Bob make a combined mea- is no contradiction here with the prohibition on surement in Bell’s basis on B and C particles. As cloning, since the initial state of the qubit was a result, A and D particles become entangled even completely erased in Alice’s laboratory and then though they never interacted with each other. recreated, but not known in Bob’s laboratory. It Note that this is equivalent to teleporting entan- should be finally stressed that here no informa- glement of one EPR pair through the other. Soon tion about the unknown state φ is transferred via the entanglement swapping was generalised to a classical channel that only conveys the message multiparticle systems [162]. It provided the opera- about the recovery operation at Bob’s lab which tional foundations of multi-photon interferometry, is completely independent on φ. in particular the method of interference of photon Original teleportation protocol was extended pairs from independent sources (see review [163]).

10 The entanglement swapping [164, 165, 166] has a hyperplane in the space of operators separating found applications among other in the generation %ent from the set of separable states S. Such a hy- of multi-photon entangled states [167], device- perplane is defined uniquely by the Hermitian op- independent key distribution [168] and construc- erator W (entanglement witness) [179]. Then the tion of quantum repeaters [169, 170, 171], quan- state is entangled iff expectation value W on %ent tum photonic [172], secret sharing [173, 174]. is negative i.e. hW i%ent < 0 whereas its expectation value on all separable states hW i%sep ≥ 0 (see Fig. 3). It was shown, that such a witness 13 Detection of quantum entanglement can be optimized by shifting the hyperplane parallel to the set S [180, 181]. Thus the detection All of the above effects and many other non- of entanglement consists in measuring the mean classical tasks based on quantum information pro- value of a properly selected observable. Remark- cessing require high purity quantum entangle- ably there is a “footbridge” Jamiolkowski isomor- ment. Unfortunately, this subtle resource is ex- phism [182] which allow to go from nonphysical tremely sensitive to interaction with the environ- positive maps to the physical measurable quan- ment and it degrades very quickly, i.e. pure states tities to Hermitian operators (entanglement wit- change into mixed (noisy) states with less entan- ness), which provides a necessary and sufficient glement. This opened up important issues: how condition separability [178]. to theoretically check whether a given state is entanglement and is it possible to detect noisy en- The entanglement witness criterion has a num- tanglement in the laboratory? ber of advantages: i) it is universal in the sense that for any entangled state always exist entan- In general, characterizing entangled states re- glement witness; ii) It certifies entanglement in gardless of the measure of utility for specific tasks experiments in the presence of noise; iii) It al- is so-called NP difficult problem [175]. The par- lows to detect the presence of entanglement even tial characterization was achieved using criteria in several measurements in contrast to tomogra- that provide the necessary but not sufficient con- phy, where the number of measurements increases ditions for deciding whether a state is entangled exponentially with the number of particles. The or not. The breakthrough was the paper of Peres disadvantage is that the witness must be pre- [176], who proposed an extremely strong separa- cisely selected for the examined state. The quan- bility test based on the partial transposition oper- tum entanglement detection based on entangle- ation. From mathematical point of view it is pos- ment witnesses has found wide applications for itive but not completely positive map thus non- the certification of two- and multi-partite states physical one. Such an operation is performed on [183, 184, 185, 186, 187, 188, 189, 190, 191] in dif- one S1 or S2 of subsystem on complex state of ferent physical scenarios. Interestingly, the con- the system S. If the state subjected to such non- cept of measurement-device-independent entan- physical surgery does not survive in the sense that glement witness which allow one to demonstrate it will cease to be positive and lose its probabilis- entanglement of all entangled quantum states tic interpretation, then the state was entangled. with untrusted measurement apparatuses was in- Mathematically speaking, this means that its par- troduced [192]. tially transposed density matrix has at least one negative eigenvalue. Based on the complete clas- The theory of entanglement detection was de- sification of positive mappings for low dimensions veloped in different directions [36, 193]. The other [177] it was proved that the PPT condition is a separability criteria based on correlation tensor necessary and sufficient condition for the separa- was proposed [194, 195, 196] for bipartite and mul- bility of 2 × 2 and 2 × 3 systems [178] which gives tipartite scenario. Recently it has been proved a complete characterization for low-dimensional that or enhanced nonlinear realignment criterion states of systems. In general necessary and suf- [197] is equivalent to the family of linear separa- ficient, albeit non-operational, separability condi- bility criteria based on correlation tensor i.e. the tion based on positive mappings was provided [36]. family of (linear) entanglement witnesses [198]. It was also demonstrated that the separability crite- The above structural criteria based on posi- ria based of the correlation tensor are weaker than tive non-physical mappings of the quantum state, positive partial transposition criterion [199]. while strong, that they cannot be implemented in a laboratory. Fortunately, based on the geometric properties of convex sets, it was possible to formu- 14 Entanglement distillation and bound late a linear separability criterion that could be entanglement implemented physically. Namely, from the convex set theory and the Hahn-Banach theorem, it fol- After Peres discovered the entanglement criterion lows that for any entangled state %ent there exists of partial transposition, a problem arose. If the

11 W W ϱent opt 0 ⟨W ⟩ϱent <

ϱsep

0 ⟨W ⟩ϱsep≥

Figure 3: The line represents hyperplane corresponding to the entanglement witness W . All states located to the left of the hyperplane or belonging to it (in particular all separable states) provide non- negative mean value of the witness, i.e. Tr(W%sep) ≥ 0 while those located to the right are entangled states detected by the witness. Wopt is optimized entanglement witness. state was untangled, it will remain untangled after entanglement that cannot be distilled with LOCC the partial transposition operation. It was natural [205]. After 12 years, several centers simultane- to ask are there states in nature that are entangled ously created the bound entanglement in the lab- and have positive partial transposition? When oratory on the photons [206, 207], on ions [208], such states were discovered in 1997 [200] they were in liquid in NMR [209], with light in continuous treated as a mathematical singularity with no ref- variable [210] regime. erence to physical reality. At about the same time, It has been shown that the bound entangle- Bennett and colleagues were working on the prob- ment is not a rare phenomenon , since its presence lem of how to reverse the entanglement degrada- was detected in thermal spin systems [211, 212]. tion process. In 1996, they published a paper that Another surprise was that the bound entangle- played a key role in the theory of entanglement ment can be activated [213] and that a crypto- manipulation [201] (see also [202]). Namely, they graphic key can be extracted from bound entan- introduced a natural class of entanglement manip- gled states it [214]. The latter lead to the general ulation operations by experimentalists in distant paradigm for distilling classical key from quantum laboratories: the two partners can perform any states in terms of so called private bits (P-bits) local operations on their entangled particles and [215] [See experimental implementation [216])]. communicate via the classical channel (LOCC). Moreover bound entangled states can violate Bell Consequently, they introduced the entanglement inequalities [217] and can be useful in quantum distillation protocol: The partners share n copies metrology [218, 219, 220]. Another an interesting of the %AB state which contains noisy entangle- open problem is the use of bound entanglement ment. With the help of local quantum opera- states in the device-independent quantum key dis- tions and classical communication, they determine tribution [221, 222, 223]. a smaller number of m (m < n) of almost maximally entangled pairs – two-qubit singlet states − |Ψ iAB. When the protocol is optimal, the constant m/n = D is a measure of entanglement in a 15 Breaking the classical order noisy state % (distillable entanglement). Distillation protocol raised the natural ques- When analysing the structure of entangled states, tion: Can all noisy states be distilled in this way? Schrödinger noticed another peculiarity of quan- It turned out that all noisy entangled two-qubit tum correlations that astonished him, as evi- states can be distilled [203]. It was a big surprise denced by the three question marks that appear that the distillation protocol does not work for in his unpublished notes in 1932 [Note in arxiv]. the higher dimension systems [204, 31]. It turned In 1935, he makes a laconic conclusion: “Best pos- out that the environment can contaminate pure sible knowledge of a whole does not include best entanglement in such a way that it is no longer possible knowledge of its parts – and that is what possible to recover it by distillation with LOCC. keeps coming back to haunt us.” [224]. It was very Thus, the entangled states with positive partial disturbing because it meant breaking the classical transposition are non-distillable. Thus, in Nature order in complex systems. As is known in the clas- there are at least two types of noise entanglement: sical world, the measure of the randomness (dis- free, that is, distillable entanglement, and bound order) of an individual random variable X is the

12 Shannon entropy: that the negativity of quantum conditional en- X tropy had no operational significance. Let us re- H(X) = − pi log pi (17) call that at the heart of the classical Shannon i communication theory is the theorem of noise- P less coding, which says that a necessary and suf- where pi – probabilities of events, pi = 1. ficient number of bits for faithful transmission is For two random variables X and Y , the to- P equal to Shannon’s entropy H [2]. Schumacher tal Shannon entropy is H(X,Y ) = pij log pij ij showed that if in Shannon’s theory we replace and conditional entropies H(X|Y ), H(Y |X) are messages by quantum states and bits by qubits, always: then the necessary and sufficient number of qubits H(X|Y ) ≡ H(X,Y ) − H(Y ) ≥ 0,H(Y |X) ≥ 0 for faithful transmission is equal to the von Neu- (18) mann entropy S(%) [23]. Soon after Schumacher which shows that the entropy of a subsystems and Westmoreland [227] and Holevo [228] gener- H(X), H(Y ) never exceeds the total entropy of alized Shannon’s channel coding theorem. Three the system H(X,Y ). kinds of quantum channel capacities was intro- In the quantum world, the measure of quan- duced: classical, quantum and private capacity, tum disorder is the von Neumann entropy S(%) which play an important role in quantum com- defined for the state %: munication [29, 229, 230, 231, 232]. The essential difference between the last two capacities X S(%) = − Tr(% log %) = hφi|% log %|φii (19) is the following: The quantum capacity is achieved in the process which guarantees that where {φi} any complete orthogonal system in H. information in any basis stays uncorrelated from When density matrix % is diagonal, it can be re- the environment after the transfer (which may garded as a quantum counterpart of a classical be shown to be equivalent to BB84 paradigm). discrete probability distribution as a natural de- Remarkably in the definition private capacity scription of quantum information source. Then much more relaxed condition is required: only one von Neumann entropy can be written in a form base is needed to stay uncorrelated in the similar to the Shannon entropy above sense. Note that the private capacity while X in general higher than the S(%) = − pi log pi, (20) quantum one may have subject to severe restric- i tion in quantum repeater scenario [233] (see more [28]). where the quantum probabilities pi are the eigen- P values of the operator % satisfy pi = 1. Meanwhile, for a long time there was no quan- The Schrödinger observation was quantified tum counterpart of Slepian-Wolf theorem [234]. using the von Neumann entropy [225, 226]. It has Namely in 1973 Slepian and Wolf formulated in been proved that the entropy of the subsystem A framework of classical communication the follow- or B can be greater than the entropy of the entire ing problem: The two partners Alice and Bob system AB only when the system is in a entan- have random variables X and Y that are corre- gled state. This implies that quantum conditional lated with each other. Bob is given some incom- entropies S(A|B) ≡ S(AB) − S(B), S(B|A) can plete information of Y in advance. Alice is in pos- be negative, which means that the disorder in the session of the missing information of X. Bob’s whole AB system may be smaller than in the sub- job is to obtain the missing information of X. systems A or B. Recalling our example with pho- The question is how much additional information tons entangled in polarization, we can see that Alice has to send to her partner. Slepian and everything happens agrees. The polarizations of Wolf proved that the amount of information that the photons measured in the laboratories of Alice Bob needs is expressed by the conditional entropy: and Bob are completely random, while the entan- H(X|Y ) ≡ H(XY ) − H(Y ) which is a measure of gled pair is in perfect order. Thus entanglement the partial information that Alice must send to can break the classical order which is the source Bob. This quantity is always positive. of the informational “paradox” of Schrödinger. In 2005, Horodecki et al. [235] proposed a quantum version of the above scenario: Alice and 16 Negative information in quantum com- Bob have a system in some unknown quantum munication state %AB which contains the complete information. Bob has some information about state %B, The breaking of the classical order was both in- while Alice has the missing information %A. The triguing and incomprehensible, especially in the task is as follows: how much information does context of Shannon’s theory, in view of the fact Alice have to send to Bob for him to have com-

13 plete information. The quantum equivalent of the first experimental measurement of a non-linear en- 2 Slepian Wolf theorem says that this quantity is tanglement witness S2(%) = − Tr ln % , using local given by the von Neumann quantum conditional measurement on two pairs of polarization entan- entropy: gled photons. At first, it seemed that the entropy criterion S(A|B) ≡ S(AB) − S(B) (21) based on nonlinear entanglement witnesses, generally weaker than the criterion based on linear ones, S(B) where is the entropy of the Bob state while will not play a major role. However, it turned out S(AB) % is the entropy of the cumulative AB that, the feature of non-linearity is its strength. In state. Contrary to the classical conditional en- particular, the nonlinear entanglement witnesses H(X|Y ) tropy , the conditional entropy can be “feel” the subtle features of entanglement in quan- both positive and negative. Conditional quantum tum multi-body systems. In last decade there has entropy has an operational interpretation of miss- been a renaissance of entropic witnesses opening S(A|B) ing information: If is positive – this is up the field for wide applications. For pure or the missing information that Alice must send to nearly pure states, entanglement was detected us- Bob via qubits (classical analogue). If S(A|B) ing Rényi S2 entropy via a multi-body quantum negative, Alice does not need to send the missing interference [241, 242, 243, 244, 245] and local ran- information via qubits. Additionally, Bob and Al- dom measurements [246, 247, 248, 249, 250]. An ice get free “quantum impulses” to send a certain experimental measurement of nonlinear witnesses number of qubits in the future, for example for of collective entanglement using hyper-entangled teleportation. two-quart states has been performed [251], see Finally it should be stressed that the above also [252]. Quite recently, an experimental multi- analysis is a strong completion of the previous re- body mixed state detection method has been pro- sult [236] which says that for any state with the posed based on the positive partial transposition quantity (21) negative there exists an entangle- of a density matrix condition. This protocol gives ment distillation protocol with one way classical the first direct PT measurement of moments in a communication (from Alice to Bob) that achieves multi-body system [253]. the number of e-bits per input noisy pair given by (21). 18 Quantum parallelism as the basis for 17 Entropy inequalities – nonlinear wit- quantum computing nesses of entanglement Quantum computing is processing information Von Neumann entropy can be generalised to the using sequence of unitary operations (quantum Rényi family α-entropy Sα(%) gates) in order to obtain an answer to a prede- termined question, e.g. is a given number factor- 1 α Sα(%) = ln Tr % , α > 1 (22) izable with high probability [254]. As we have 1 − α seen single qubit allows two basic states to be stored and processed simultaneously. The prob- It is easy to check that the Rényi entropy in lem is that the decoherence process being a re- the α → 1 limit turns into the von Neumann en- sult of disturbance by environment occurs within tropy S(%). The natural question was whether a short time (decoherence time) destroys coher- there are quantum states that satisfy the analog ence. Roughly speaking decoherence time is the of classical inequalities (18). In 1996 [237] it was characteristic time for a generic qubit state (2) to proved that all non-entangled (separable) states be transformed into the mixture % = |a|2|0ih0| + at a finite dimensional Hilbert space for α = 1.2 |b|2|1ih1|. One of the basic conditions for effec- satisfy α-entropic inequalities: tive quantum computing requires that long rel-

Sα(A|B) = Sα(%AB) − Sα(%B) ≥ 0, evant decoherence times, much longer than the gate operation time. This is one of the ﬁve basic Sα(B|A) = Sα(%AB) − Sα(%A) ≥ 0 (23) DiVincenzo criteria required for a physical implementation of quantum computing [255]. If we take It presents entropic nonlinear entanglement a superposition of n qubits then a pure state will criterion which does not require a priori knowl- represent a simultaneous superposition of N = 2n edge of the state. possible distinct basic states. Nonlinear experimentally friendly collective entanglement witnesses were also proposed, which N−1 also do not require prior knowledge of a given state X |Ψi = Ci|ii (24) [238, 239]. In [240] Bovino et al. demonstrated i=0

14 It is remarkable, that one can processes simul- 3. A “universal” set of quantum gates. taneously an exponential number of basic states. This feature (quantum parallelism) underlies the 4. A qubit-specific measurement capability. superiority of quantum computing over classical Notoriously the quantum computing process is one. To illustrate the latter suppose that we disturbed by the interaction with the environ- have access to quantum oracle that computes a ment, causing the occurrence of errors. Therefore given function f(i) from an input i of n qubits both bit (0,1) and phase (“0 + 1”, “0 − 1”) they n (i = 0, 1 ... 2 ). must be protected. This seems impossible due to Having a prepared string of qubits in the fidu- the non-cloning theorem. Fortunately, Shor [21] cial state of 0 and applying to each qubit, in par- and Steane [262] overcame this difficulty by intro- allel, Hadamard gate, we obtain a register of n duction of the error correction codes. The trick qubits in an equal superposition of all bit strings is that the information of a one logical qubit can N−1 be spread onto a highly entangled state of several 1 X H|0i ⊗ H|0i ⊗ · · · ⊗ H|0i = √ |ii (25) physical qubits. N i=0 |0i → |0i = [(|000i + |111i)(|000i + |111i) L √ where |ii is the computational basis state indexed ×(|000i + |111i)]/2 2 (27) by the binary number that would correspond to |1i → |1iL = [(|000i − |111i)(|000i − |111i) the number i in base-10 notation. √ Now suppose that the function f is evaluated ×(|000i − |111i))]/2 2 (28) by unitary transformation Uf : |xi|0i → |xi|f(x)i. This code first introduced by Shor [21] corrects Then the linearity of quantum formalism implies both bit error σx and phase error σz. N−1 N−1 Of course the error correction procedure itself 1 X 1 X Uf : √ |ii|0i → √ |ii|f(0)i is not error-free. Fortunately the possibility of N N i=0 i=0 efficient quantum computing is based on the so- (26) called the threshold theorem: Error probability of This means that all possible evaluations of the elementary operation smaller than some thresh- function f(x) can be evaluated in a single step. old value p < p0 then efficient quantum comput- The idea of quantum computing received a lot ing possible [263, 264]. In practice, this condition, of support when it was discovered that certain which is the basis of efficient quantum computing, difficult computational problems such as number is extremely demanding. Already in 1995 it was factoring (Shor’s algorithm [22]) and searching un- demonstrated that the quantum computing can structured data (Grover’s algorithm [24]) can be be implemented with cold ions confined in linear solved far more efficiently than classically. The ef- trap and interacting laser beams [265]. The first ficiency of computation is measured by the com- 7-qubit quantum computer from IBM and Stan- putation complexity that is number of steps re- ford University based on nuclear magnetic reso- quired to solve a given task as a function of the size nance realized Shor’s algorithm, decomposition of of the input. The important algorithms: Deutsch- the number 15 = 3x5 [256]. The scale of the diffi- Jozsa [18], Shor [22] and Grover [24] have been culties is evidenced by the fact that a qualitative discovered that demonstrate quantum supremacy breakthrough in this field took place only after 18 over classical computing. All three algorithms years. Namely researchers at Google’s quantum- have be implemented on primitive quantum com- computing laboratory in Santa Barbara, Califor- puters based on NMR techniques [256], in ions nia, announced the first-ever demonstration of traps [257] and quantum dots [258]. Since then quantum supremacy on the 53 qubit quantum many other algorithms have been discovered, such computer Sycomore, made of superconducting cir- as quantum simulations, [259] variational quan- cuits that are kept at ultracold temperatures [266]. tum solvers [260] which demonstrate quantum It executes algorithms quantum with 1500 gates. supremacy [see more [261]]. It is impressive achievement, however, it was de- Any realistic implementation of universal signed for a specific problem — boson sampling quantum computation is big challenge. It must [267], which is simplified non-universal model for meet the DiVincenzo criteria [255]. Except deco- quantum computing that may hold the key to im- herence criterion, there are four more: plementing the first ever post-classical quantum 1. A scalable physical system with well charac- computer. More specifically this is the process in terized qubits. which a very nontrivial output statistic is achieved quantumly which requires (under some reasonable 2. The ability to initialize the state of the assumptions) exponentially longer time to be pro- qubits to a simple fiducial state. duced by classical machines. While it is not a

15 quantum algorithm in a standard form its remark- where H is the generator of the phase shift of the P P able practical application to fast finding of some system, and % = k λk|kihk|, k λk = 1. graph properties are predicted. For unitary dynamics of the linear two-mode Quite recently Jian-Wei Pan and colleagues interferometer the generator of the phase shift is at the University of Science and Technology of H ≡ J~n = ~n · J where J~n is a component of China in Hefei et al. announced in December 2020 the collective spin operator angular momentum photon-based quantum computer, which demon- in the direction ~n. It has been shown [272, 273], strates quantum supremacy via boson-sampling that for the separable input N-particle states, with 50-70 detected photons [268]. It could find the quantum Fisher information is bounded by solutions to the boson-sampling problem in 200 FQ[%sep, J~n] ≤ N. Hence the phase uncertainty seconds, while classical China’s Taihu-Light su- ∆θ˜ is bounded by standard quantum limit (SQL) ˜ percomputer. need 2.5 billion years. In in con- ∆θSN : ∆θ ≥ ∆θSN where trast to Google’s Sycamore, the Chinese team’s 1 photonic circuit is not programmable [269]. ∆θSN = √ (32) mN 19 Entanglement – resource in quantum By using entangled probe states it is possible to metrology overcome the SQL [49]. Quantum formalism imposes fundamental constraints on measurement The discovery that the use of entangled states in precision that scales like 1/N. It has been shown quantum metrology can improve the precision of that, for general probe states of N particles FQ measurements [270, 271] led to the rapid develop- 2 is bounded by FQ[%, J~n] ≤ N , [272, 273] and ment of quantum enhanced metrology [49] which this inequality can be saturated by certain maxi- allows measure physical quantities by estimating mally entangled states. It allows to obtain optimal θ the phase shift using interferometric techniques. Heisenberg bound for the phase uncertainty A basic problem in quantum metrology can be formulated as in the diagram (Fig. 4): A probe state 1 ∆θHN = √ (33) % of N particles is prepared and then subject to a mN parameter-dependent quantum channel Λθ. The state %θ = Λθ(%) is finally measured via POVM Note that the genuine multipartite entanglement measurement {Πi}I . It provides conditional prob- is needed for reaching the highest sensitivities in ability distribution p(i|θ), which is used to esti- some metrological tasks using two-mode linear in- mate of θ via estimator function Θ(˜ i) for given terferometer [274, 275, 276]. Recently, various the measurement outcome i. The estimation of experiments have demonstrated beating the SQL the phase shift is limited by uncertainty: (see [277] and references there in). In a realistic scenario, quantum phase estima- ∆2θ˜ = h(θ˜ − θ)2i (29) tion requires taking into account the influence effects of losses and decoherence [278, 279, 280, 281, h i where means the average over all measurement 282, 283, 284, 285]. In particular for N probe results. The task is to find the optimal probe particles prepared in state %N and noisy chan- state %, the optimal measurement Π and estima- ⊗N nel ΛΘ , that acts independently on each par- tor , which minimize the uncertainty. For unbi- N ⊗N N ticle %θ = ΛΘ (% ), quantum Fisher informa- ased estimators and m independent measurements N tion FQ(%θ ) has asymptotically in N a bound the phase uncertainty is limited by the quantum N that scales linearly with N: FQ(%θ ) ≤ Nα giving Cramer-Rao bound: bound [281]: 1 ˜ 1 ∆θ˜ ≥ √ , ∆θ ≥ p (30) (34) mFQ(%θ) αmN where α is constant. Thus the supremacy over where FQ is quantum Fisher information which SQL is only limited to constants factor. In partic- quantifies asymptotic usefulness of quantum state ular, in the optical interferometry with losses for a and it can be estimated for the different quantum generic two mode input N-photon state with pre- channels [49]. cisely defined total photon number N the limit of For unitary and noiseless quantum channel −iHθ +iHθ phase sensitivity is: %θ = Λθ(%) = e % e the quantum Fisher information optimized over measurement can be r1 − η ∆θ˜ ≥ (35) expressed in the form: ηN 2 X (λk − λl) 2 η FQ[%, H] = 2 |hk|H|li| (31) where is optical transfer coefficient. This bound λk + λl k,l generalized to states having uncertainty photon

16 state quantum measurement estimation preparation channel Π ∼ Λθ i θ(i)

Figure 4: Phase estimation scheme number such as coherent states and squeezed 20 Final remarks states was used to estimate the fundamental bound on GEO 600 interferometer strain sensitiv- In this article, I have focused only on selected as- ity [286] where the phase noise decoherence [287], pects of quantum information. There are many and quantum back-action are negligible [270]. It other fascinating phenomena that deserve presen- has been shown that the coherent-state squeezed tation. These include quantum correlations be- vacuum strategy is optimal one for phase estima- yond entanglement [44, 291], nonlocality with- tion with high precision on absolute scale [286]. out entanglement [292], quantum channel super Recently, a framework for optimization of activation effect [293, 44], locking classical cor- quantum metrological protocols based on the ten- relations in quantum states [294], resources the- sor network approach for the channel with the cor- oretical approach to quantum thermodynamics related noise and the phase parameter unitarily [40], quantum Darwinism [295, 296, 297], objec- encoded were presented [285]. Note that multipa- tivity [298, 299, 300, 301], quantum based ran- rameter estimation theory offers a general frame- domness amplification against postquantum at- work to explore imaging techniques beyond the tacks [302, 303, 304] and others. They all under- Rayleigh limit [288]. line the extremely complex nature of quantum in- Overall, the relationship between quantum formation, which is not yet fully understood and metrology and the structure of quantum states is provokes many open questions (see for example still not entirely clear. For example there are very [305]). Among others there is a long-standing weakly entangled states (bound entangled states) question: If the quantum formalism can be con- metrologically useful [218, 219] as well highly en- sistently extend to include quantum gravitation tangled states that are not useful for metrology effect? If so, how it will impact on the quantum [289]. It leads to the question: Are there situ- information concept? ations were some synergy effects occurs possibly with analogy to previous communication proto- Acknowledgments cols such as activation of bound entanglement? In attempt to answer this question, the criterion I would like to thank J. Mostowski, and A. Wys- of metrological usefulness have been proposed as mołek for encouraging me to write this article follows [290]: based on the lecture given at the Extraordinary % The state is metrologically useful iff there Congress of Polish Physicists on the occasion of H exists Hamiltonian such that Fisher quantum the centenary of the Polish Physical Society. I information (31) is sharply greater than Fisher in- would like also to thank J. Horodecka and Ł. F [% ,H] formation for separable states Q sep maxi- Pankowski for their help in editing of this paper. I mized over all separable states: acknowledge support by the Foundation for Polish (sep) Science through IRAP project cofinanced by the FQ[%, H] > max = FQ[%sep,H] =: FQ (H) %sep EU within the Smart Growth Operational Pro- (36) gramme (Contract No. 2018/MAB/5). Then the metrological gain with respect to the Hamiltonian H defines as gH (%) = (sep) FQ[%, H]/FQ (H) leads to the optimal gain References g(%) = maxlocalH gH (%). Having such defined metrological usefulness it has been shown that the [1] J. von Neumann. Mathematische Grundla- bipartite entangled states that cannot outperform gen der Quantenmechanic. Springer, Berlin separable states in any linear interferometer, how- (1932). ever can still be more useful than separable ones if several copies of them are considered or an ancilla [2] C. E. Shannon. A Mathematical Theory is added to the quantum system. In particular it of Communication. Bell System Techni- has been proved that all entangled bipartite pure cal Journal 27, 379–423 (1948). DOI: states are metrologically useful. 10.1002/j.1538-7305.1948.tb01338.x.

17 [3] A. Einstein, B. Podolsky, and N. Rosen. [14] D. Deutsch. Quantum theory, the Can Quantum-Mechanical Description of Church–Turing principle and the universal Physical Reality Be Considered Complete? quantum computer. Proceedings of the Royal Physical Review 47, 777–780 (1935). DOI: Society of London. A. Mathematical and 10.1103/physrev.47.777. Physical Sciences 400, 97–117 (1985). DOI: 10.1098/rspa.1985.0070. [4] J. L. Park. The concept of transition in quantum mechanics. Foundations of Physics [15] R. P. Feynman. Quantum mechanical com- 1, 23–33 (1970). DOI: 10.1007/bf00708652. puters. Foundations of Physics 16, 507–531 (1986). DOI: 10.1007/bf01886518. [5] S. Wiesner. Conjugate coding. ACM SIGACT News 15, 78–88 (1983). DOI: [16] A. K. Ekert. Quantum cryptography based 10.1145/1008908.1008920. on Bell’s theorem. Physical Review Letters 67, 661–663 (1991). DOI: 10.1103/phys- [6] A. S. Holevo. Bounds for the Quantity of In- revlett.67.661. formation Transmitted by a Quantum Com- [17] C. H. Bennett and S. J. Wiesner. Commu- munication Channel. Probl. Peredachi Inf. nication via one- and two-particle operators 9 , 3 (1973). on Einstein-Podolsky-Rosen states. Physical 69 [7] J. P. Gordon. Noise at optical frequencies; Review Letters , 2881–2884 (1992). DOI: information theory. In P. A. Miles, editor, 10.1103/physrevlett.69.2881. Quantum Electronics and Coherent Light; [18] D. Deutsch and R. Jozsa. Rapid solu- Proceedings of the International School of tion of problems by quantum computation. Physics Enrico Fermi, Course XXXI, 156. Proceedings of the Royal Society of Lon- Academic Press New York (1964). don. Series A: Mathematical and Physi- cal Sciences 439, 553–558 (1992). DOI: [8] L. B. Levitin. On the quantum measure of 10.1098/rspa.1992.0167. information. In Proceedings of the Fourth All-Union Conference on Information and [19] C. H. Bennett, G. Brassard, C. Crépeau, Coding Theory, Sec. II. Tashkent (1969). R. Jozsa, A. Peres, and W. K. Woot- ters. Teleporting an unknown quantum state [9] R. S. Ingarden. Quantum information via dual classical and Einstein-Podolsky- theory. Reports on Mathematical Physics Rosen channels. Physical Review Letters 10, 43–72 (1976). DOI: 10.1016/0034- 70, 1895–1899 (1993). DOI: 10.1103/phys- 4877(76)90005-7. revlett.70.1895. [10] R. Ingarden, A. Kossakowski, and M. Ohya. [20] M. Żukowski, A. Zeilinger, M. A. Horne, Information Dynamics and Open Systems: and A. K. Ekert. “Event-ready-detectors” Classical and Quantum Approach. Kluwer Bell experiment via entanglement swap- Academic Publishers (1997). ping. Physical Review Letters 71, 4287–4290 (1993). DOI: 10.1103/physrevlett.71.4287. [11] J. S. Bell. On the Einstein Podolsky Rosen paradox. Physics Physique 1, 195– [21] P. W. Shor. Scheme for reducing decoher- 200 (1964). DOI: 10.1103/physicsphysique- ence in quantum computer memory. Phys- ﬁzika.1.195. ical Review A 52, R2493–R2496 (1995). DOI: 10.1103/physreva.52.r2493. [12] A. Aspect, J. Dalibard, and G. Roger. Ex- perimental Test of Bell's Inequalities Us- [22] P. W. Shor. Polynomial-Time Algo- ing Time- Varying Analyzers. Physical Re- rithms for Prime Factorization and Dis- view Letters 49, 1804–1807 (1982). DOI: crete Logarithms on a Quantum Computer. 26 10.1103/physrevlett.49.1804. SIAM Journal on Computing , 1484–1509 (1997). DOI: 10.1137/s0097539795293172. [13] C. H. Bennett and G. Brassard. Quan- [23] B. Schumacher. Quantum coding. Physi- tum Cryptography: Public Key Distribution cal Review A 51, 2738–2747 (1995). DOI: and Coin Tossing. In Proceedings of the 10.1103/physreva.51.2738. IEEE International Conference on Comput- ers, Systems and Signal Processing, 175– [24] L. K. Grover. Quantum Mechanics Helps in 179. IEEE Computer Society Press, New Searching for a Needle in a Haystack. Physi- York, Bangalore, India, December 1984 cal Review Letters 79, 325–328 (1997). DOI: (1984). 10.1103/physrevlett.79.325.

18 [25] D. Bouwmeester, J.-W. Pan, K. Mattle, [37] E. Schrödinger. Discussion of Probabil- M. Eibl, H. Weinfurter, and A. Zeilinger. ity Relations between Separated Systems. Experimental quantum teleportation. Nature Mathematical Proceedings of the Cambridge 390, 575–579 (1997). DOI: 10.1038/37539. Philosophical Society 31, 555–563 (1935). DOI: 10.1017/s0305004100013554. [26] D. Boschi, S. Branca, F. D. Martini, L. Hardy, and S. Popescu. Experimen- [38] R. Horodecki, M. Horodecki, and tal Realization of Teleporting an Unknown P. Horodecki. Quantum information Pure Quantum State via Dual Classical and isomorphism: Beyond the dilemma of the Einstein-Podolsky-Rosen Channels. Phys- Scylla of ontology and the Charybdis of ical Review Letters 80, 1121–1125 (1998). instrumentalism. IBM Journal of Research DOI: 10.1103/physrevlett.80.1121. and Development 48, 139–147 (2004). DOI: 10.1147/rd.481.0139. [27] A. Furusawa, J. L. Sørensen, S. L. Braun- stein, C. A. Fuchs, H. J. Kimble, and E. S. [39] J. M. Raimond, M. Brune, and S. Haroche. Polzika. Unconditional Quantum Teleporta- Manipulating quantum entanglement with tion. Science 282, 706–709 (1998). DOI: atoms and photons in a cavity. Reviews of 10.1126/science.282.5389.706. Modern Physics 73, 565–582 (2001). DOI: [28] M. Wilde. From Classical to Quantum 10.1103/revmodphys.73.565. Shannon Theory. Cambridge University [40] E. Chitambar and G. Gour. Quantum re- Press (2019). source theories. Reviews of Modern Physics [29] C. H. Bennett, D. P. DiVincenzo, J. A. 91, 025001 (2019). DOI: 10.1103/revmod- Smolin, and W. K. Wootters. Mixed-state phys.91.025001. entanglement and quantum error correction. Physical Review A 54, 3824–3851 (1996). [41] L. Amico, R. Fazio, A. Osterloh, and V. Ve- DOI: 10.1103/physreva.54.3824. dral. Entanglement in many-body systems. Reviews of Modern Physics 80, 517–576 [30] V. Coﬀman, J. Kundu, and W. K. Wootters. (2008). DOI: 10.1103/revmodphys.80.517. Distributed entanglement. Physical Review A 61, 052306 (2000). DOI: 10.1103/phys- [42] O. Gühne and G. Tóth. Entanglement de- reva.61.052306. tection. Physics Reports 474, 1–75 (2009). DOI: 10.1016/j.physrep.2009.02.004. [31] B. M. Terhal, M. M. Wolf, and A. C. Do- herty. Quantum Entanglement: A Mod- [43] H. M. Wiseman, S. J. Jones, and A. C. Do- ern Perspective. Physics Today 56, 46–52 herty. Steering, Entanglement, Nonlocality, (2003). DOI: 10.1063/1.1580049. and the Einstein-Podolsky-Rosen Paradox. Physical Review Letters 98, 140402 (2007). [32] M. A. Nielsen and I. L. Chuang. Quan- DOI: 10.1103/physrevlett.98.140402. tum Computation and Quantum Informa- tion. Cambridge University Press, Cam- [44] K. Modi, A. Brodutch, H. Cable, T. Pa- bridge (2000). terek, and V. Vedral. The classical- quantum boundary for correlations: Dis- [33] H.-P. Breuer and F. Petruccione. The The- cord and related measures. Reviews of Mod- ory of Open Quantum Systems. Oxford Uni- ern Physics 84, 1655–1707 (2012). DOI: versity Press (2002). 10.1103/revmodphys.84.1655. [34] W. Heisenberg. Über den anschaulichen [45] A. Streltsov, G. Adesso, and M. B. Ple- Inhalt der quantentheoretischen Kinematik nio. Colloquium : Quantum coherence as und Mechanik. Zeitschrift für Physik 43, a resource. Reviews of Modern Physics 172–198 (1927). DOI: 10.1007/bf01397280. 89, 041003 (2017). DOI: 10.1103/revmod- [35] N. Bohr. New Problems in Quantum The- phys.89.041003. ory. Nature 121, 579–579 (1928). DOI: 10.1038/121579a0. [46] J. Goold, M. Huber, A. Riera, L. del Rio, and P. Skrzypczyk. The role of quantum in- [36] R. Horodecki, P. Horodecki, M. Horodecki, formation in thermodynamics—a topical re- and K. Horodecki. Quantum entanglement. view. Journal of Physics A: Mathematical Reviews of Modern Physics 81, 865–942 and Theoretical 49, 143001 (2016). DOI: (2009). DOI: 10.1103/revmodphys.81.865. 10.1088/1751-8113/49/14/143001.

19 [47] V. Scarani, A. Acín, G. Ribordy, and L. P. Kouwenhoven. Electron spin qubits in N. Gisin. Quantum Cryptography Proto- quantum dots. In IEDM Technical Digest. cols Robust against Photon Number Split- IEEE International Electron Devices Meet- ting Attacks for Weak Laser Pulse Im- ing, 2004, 533. IEEE (2005). plementations. Physical Review Letters 92, 057901 (2004). DOI: 10.1103/phys- [57] D. Press, T. D. Ladd, B. Zhang, and Y. Ya- revlett.92.057901. mamoto. Complete quantum control of a single quantum dot spin using ultrafast op- [48] N. Gisin, G. Ribordy, W. Tittel, and tical pulses. Nature 456, 218–221 (2008). H. Zbinden. Quantum cryptography. Re- DOI: 10.1038/nature07530. views of Modern Physics 74, 145–195 (2002). DOI: 10.1103/revmodphys.74.145. [58] T. Yamamoto, Y. A. Pashkin, O. Astafiev, Y. Nakamura, and J. S. Tsai. Demonstra- [49] L. Pezzè, A. Smerzi, M. K. Oberthaler, tion of conditional gate operation using su- R. Schmied, and P. Treutlein. Quantum perconducting charge qubits. Nature 425, metrology with nonclassical states of atomic 941–944 (2003). DOI: 10.1038/nature02015. ensembles. Reviews of Modern Physics 90, 035005 (2018). DOI: 10.1103/revmod- [59] A. F. Kockum and F. Nori. Quantum phys.90.035005. Bits with Josephson Junctions. In Funda- mentals and Frontiers of the Josephson Ef- [50] H. Buhrman, R. Cleve, S. Massar, and fect, 703–741. Springer International Pub- R. de Wolf. Nonlocality and communica- lishing (2019). DOI: 10.1007/978-3-030- tion complexity. Reviews of Modern Physics 20726-7_17. 82, 665–698 (2010). DOI: 10.1103/revmodphys.82.665. [60] E. Knill, R. Laflamme, and G. J. Milburn. A scheme for efficient quantum computa- [51] Č. Brukner, M. Żukowski, J.-W. Pan, and tion with linear optics. Nature 409, 46–52 A. Zeilinger. Bell’s Inequalities and Quan- (2001). DOI: 10.1038/35051009. tum Communication Complexity. Physical Review Letters 92, 127901 (2004). DOI: [61] P. Kok, W. J. Munro, K. Nemoto, T. C. 10.1103/physrevlett.92.127901. Ralph, J. P. Dowling, and G. J. Milburn. Linear optical quantum computing with pho- [52] P. Trojek, C. Schmid, M. Bourennane, tonic qubits. Reviews of Modern Physics Č. Brukner, M. Żukowski, and H. We- 79, 135–174 (2007). DOI: 10.1103/revmod- infurter. Experimental quantum commu- phys.79.135. nication complexity. Physical Review A 72, 050305 (2005). DOI: 10.1103/phys- [62] M. Parniak, M. Dąbrowski, M. Maze- reva.72.050305. lanik, A. Leszczyński, M. Lipka, and W. Wasilewski. Wavevector multiplexed [53] W. H. Zurek. Decoherence, einselection, atomic quantum memory via spatially- and the quantum origins of the classical. resolved single-photon detection. Nature Reviews of Modern Physics 75, 715–775 Communications 8, 2140 (2017). DOI: (2003). DOI: 10.1103/revmodphys.75.715. 10.1038/s41467-017-02366-7.

[54] R. Heilmann, M. Gräfe, S. Nolte, and [63] E. P. Wigner. The Probability of the Ex- A. Szameit. Arbitrary photonic wave plate istence of a Self-Reproducing Unit London. operations on chip: Realizing Hadamard, (1961). Pauli-X and rotation gates for polarisation qubits. Scientiﬁc Reports 4, 4118 (2014). [64] W. K. Wootters and W. H. Zurek. A sin- DOI: 10.1038/srep04118. gle quantum cannot be cloned. Nature 299, 802–803 (1982). DOI: 10.1038/299802a0. [55] V. V. Samartsev and T. G. Mitrofanova. Qubits based on the exciton degrees of [65] D. Dieks. Communication by EPR devices. freedom of a semiconductor quantum dot. Physics Letters A 92, 271–272 (1982). DOI: Journal of Physics: Conference Series 10.1016/0375-9601(82)90084-6. 1283, 012012 (2019). DOI: 10.1088/1742- 6596/1283/1/012012. [66] V. Bužek and M. Hillery. Quantum copying: Beyond the no-cloning theorem. Phys- [56] R. Hanson, J. M. Elzerman, L. H. W. ical Review A 54, 1844–1852 (1996). DOI: van Beveren, L. M. K. Vandersypen, and 10.1103/physreva.54.1844.

20 [67] H. Fan, Y.-N. Wang, L. Jing, J.-D. Yue, H.- Optic Engineers Conference. OSA (2009). D. Shi, Y.-L. Zhang, and L.-Z. Mu. Quan- DOI: 10.1364/ofc.2009.othl2. tum cloning machines and the applications. Physics Reports 544, 241–322 (2014). DOI: [77] B. Korzh, C. C. W. Lim, R. Houlmann, 10.1016/j.physrep.2014.06.004. N. Gisin, M. J. Li, D. Nolan, B. Sanguinetti, R. Thew, et al. Provably secure and practi- [68] A. K. Pati and S. L. Braunstein. Im- cal quantum key distribution over 307 km of possibility of deleting an unknown quantum optical fibre. Nature Photonics 9, 163–168 404 state. Nature , 164–165 (2000). DOI: (2015). DOI: 10.1038/nphoton.2014.327. 10.1038/404130b0. [69] U. Alvarez-Rodriguez, M. Sanz, L. Lamata, [78] S.-K. Liao, W.-Q. Cai, J. Handsteiner, and E. Solano. The Forbidden Quantum B. Liu, J. Yin, L. Zhang, D. Rauch, M. Fink, Adder. Scientific Reports 5, 11983 (2015). et al. Satellite-Relayed Intercontinental DOI: 10.1038/srep11983. Quantum Network. Physical Review Letters 120, 030501 (2018). DOI: 10.1103/phys- [70] M. Oszmaniec, A. Grudka, M. Horodecki, revlett.120.030501. and A. Wójcik. Creating a Superposition of Unknown Quantum States. Physical Re- [79] W. O. Krawec, R. Liss, and T. Mor. Se- view Letters 116, 110403 (2016). DOI: curity Proof Against Collective Attacks for 10.1103/physrevlett.116.110403. an Experimentally Feasible Semi-Quantum Key Distribution Protocol. arXiv e-prints, [71] K. Li, G. Long, H. Katiyar, T. Xin, G. Feng, arXiv:2012.02127 (2020). D. Lu, and R. Laflamme. Experimentally su- perposing two pure states with partial prior [80] D. M. Greenberger, M. A. Horne, A. Shi- 95 knowledge. Physical Review A , 022334 mony, and A. Zeilinger. Bell’s theorem (2017). DOI: 10.1103/physreva.95.022334. without inequalities. American Journal [72] X.-M. Hu, M.-J. Hu, J.-S. Chen, B.-H. Liu, of Physics 58, 1131–1143 (1990). DOI: Y.-F. Huang, C.-F. Li, G.-C. Guo, and Y.-S. 10.1119/1.16243. Zhang. Experimental creation of superposition of unknown photonic quantum states. [81] W. Dür, G. Vidal, and J. I. Cirac. Three Physical Review A 94, 033844 (2016). DOI: qubits can be entangled in two inequivalent 62 10.1103/physreva.94.033844. ways. Physical Review A , 062314 (2000). DOI: 10.1103/physreva.62.062314. [73] C. H. Bennett, F. Bessette, G. Bras- sard, L. Salvail, and J. Smolin. Ex- [82] W.-X. Cui, S. Hu, H.-F. Wang, A.-D. Zhu, perimental quantum cryptography. Jour- and S. Zhang. Deterministic conversion of a nal of Cryptology 5, 3–28 (1992). DOI: four-photon GHZ state to a W state via ho- 10.1007/bf00191318. modyne measurement. Optics Express 24, 15319 (2016). DOI: 10.1364/oe.24.015319. [74] D. Stucki, N. Gisin, O. Guinnard, G. Ri- bordy, and H. Zbinden. Quantum key distri- [83] T. Haase, G. Alber, and V. M. Stojanovic. bution over 67 km with a plug&play system. W-to-GHZ state conversion in the Rydberg- 4 New Journal of Physics , 41–41 (2002). blockade regime of neutral-atom systems: DOI: 10.1088/1367-2630/4/1/341. dynamical-symmetry-based approach. arXiv [75] S. Wengerowsky, S. K. Joshi, F. Stein- e-prints, arXiv:2012.15169 (2020). lechner, J. R. Zichi, S. M. Dobrovol- skiy, R. van der Molen, J. W. N. Los, [84] F. Benatti, R. Floreanini, F. Fran- V. Zwiller, et al. Entanglement distribu- chini, and U. Marzolino. Entangle- tion over a 96-km-long submarine optical ment in indistinguishable particle systems. 878 fiber. Proceedings of the National Academy Physics Reports , 1–27 (2020). DOI: of Sciences 116, 6684–6688 (2019). DOI: 10.1016/j.physrep.2020.07.003. 10.1073/pnas.1818752116. [85] R. Y. Chiao, P. G. Kwia, and A. M. Stein- [76] M. Peev, T. Länger, T. Lorünser, A. Happe, berg. Quantum non-locality in two-photon O. Maurhart, A. Poppe, and T. Themel. experiments at Berkeley. Quantum and The SECOQC Quantum-Key-Distribution Semiclassical Optics: Journal of the Eu- Network in Vienna. In Optical Fiber Com- ropean Optical Society Part B 7, 259–278 munication Conference and National Fiber (1995). DOI: 10.1088/1355-5111/7/3/006.

21 [86] M. Erhard, M. Krenn, and A. Zeilinger. Ad- [97] B. S. Cirel'son. Quantum generalizations vances in high-dimensional quantum entan- of Bell's inequality. Letters in Mathe- glement. Nature Reviews Physics 2, 365–381 matical Physics 4, 93–100 (1980). DOI: (2020). DOI: 10.1038/s42254-020-0193-5. 10.1007/bf00417500. [87] N. Akopian, N. H. Lindner, E. Poem, [98] D. I. Kaiser. Tackling Loopholes in Exper- Y. Berlatzky, J. Avron, D. Gershoni, imental Tests of Bell’s Inequality. arXiv e- B. D. Gerardot, and P. M. Petroff. En- prints, arXiv:2011.09296 (2020). tangled Photon Pairs from Semiconductor [99] T. Scheidl, R. Ursin, J. Kofler, S. Ramelow, Quantum Dots. Physical Review Letters X.-S. Ma, T. Herbst, L. Ratschbacher, 96, 130501 (2006). DOI: 10.1103/phys- A. Fedrizzi, et al. Violation of local revlett.96.130501. realism with freedom of choice. Pro- [88] B. Yurke and D. Stoler. Einstein-Podolsky- ceedings of the National Academy of Sci- Rosen effects from independent particle ences 107, 19708–19713 (2010). DOI: sources. Physical Review Letters 68, 10.1073/pnas.1002780107. 1251–1254 (1992). DOI: 10.1103/phys- [100] D. Aktas, S. Tanzilli, A. Martin, G. Pütz, revlett.68.1251. R. Thew, and N. Gisin. Demonstration [89] K. Życzkowski, P. Horodecki, M. Horodecki, of Quantum Nonlocality in the Presence and R. Horodecki. Dynamics of quan- of Measurement Dependence. Physical Re- tum entanglement. Physical Review A view Letters 114, 220404 (2015). DOI: 65, 012101 (2001). DOI: 10.1103/phys- 10.1103/physrevlett.114.220404. reva.65.012101. [101] M.-H. Li, C. Wu, Y. Zhang, W.-Z. Liu, [90] T. Yu and J. Eberly. Sudden death of en- B. Bai, Y. Liu, W. Zhang, Q. Zhao, et al. tanglement: Classical noise effects. Op- Test of Local Realism into the Past without tics Communications 264, 393–397 (2006). Detection and Locality Loopholes. Physical DOI: 10.1016/j.optcom.2006.01.061. Review Letters 121, 080404 (2018). DOI: 10.1103/physrevlett.121.080404. [91] T. Yu and J. H. Eberly. Sudden Death of En- tanglement. Science 323, 598–601 (2009). [102] M. Żukowski and Č. Brukner. Quan- DOI: 10.1126/science.1167343. tum non-locality—it ain’t necessarily so... Journal of Physics A: Mathematical and [92] K. Życzkowski, P. Horodecki, A. Sanpera, Theoretical 47, 424009 (2014). DOI: and M. Lewenstein. Volume of the set of 10.1088/1751-8113/47/42/424009. separable states. Physical Review A 58, 883– 892 (1998). DOI: 10.1103/physreva.58.883. [103] E. G. Cavalcanti and H. M. Wiseman. Bell Nonlocality, Signal Locality and Un- [93] M. Kuś and K. Życzkowski. Geometry predictability (or What Bohr Could Have of entangled states. Physical Review A Told Einstein at Solvay Had He Known 63, 032307 (2001). DOI: 10.1103/phys- About Bell Experiments). Foundations reva.63.032307. of Physics 42, 1329–1338 (2012). DOI: [94] P. G. Kwiat, K. Mattle, H. Weinfurter, 10.1007/s10701-012-9669-1. A. Zeilinger, A. V. Sergienko, and Y. Shih. [104] S. Pironio, A. Acín, N. Brunner, N. Gisin, New High-Intensity Source of Polarization- S. Massar, and V. Scarani. Device- Entangled Photon Pairs. Physical Re- independent quantum key distribution se- view Letters 75, 4337–4341 (1995). DOI: cure against collective attacks. New Jour- 10.1103/physrevlett.75.4337. nal of Physics 11, 045021 (2009). DOI: 10.1088/1367-2630/11/4/045021. [95] S. L. Braunstein, A. Mann, and M. Revzen. Maximal violation of Bell inequalities for [105] N. Brunner, D. Cavalcanti, S. Pironio, mixed states. Physical Review Letters 68, V. Scarani, and S. Wehner. Bell nonlocal- 3259–3261 (1992). DOI: 10.1103/phys- ity. Reviews of Modern Physics 86, 419–478 revlett.68.3259. (2014). DOI: 10.1103/revmodphys.86.419. [96] J. F. Clauser, M. A. Horne, A. Shimony, [106] R. Arnon-Friedman and J.-D. Bancal. and R. A. Holt. Proposed Experiment to Device-independent certification of one-shot Test Local Hidden-Variable Theories. Physi- distillable entanglement. New Journal cal Review Letters 23, 880–884 (1969). DOI: of Physics 21, 033010 (2019). DOI: 10.1103/physrevlett.23.880. 10.1088/1367-2630/aafef6.

22 [107] M. Farkas, M. Balanzó-Juandó, [117] N. Miklin, J. J. Borkała, and M. Pawłowski. K. Łukanowski, J. Kołodyński, and A. Acín. Semi-device-independent self-testing of Bell nonlocality is not suﬃcient for the unsharp measurements. Physical Re- security of standard device-independent view Research 2, 033014 (2020). DOI: quantum key distribution protocols. arXiv 10.1103/physrevresearch.2.033014. e-prints, arXiv:2103.02639v1 (2021). [118] R. Ramanathan, D. Goyeneche, S. Muham- [108] S. Kochen and E. Specker. The Prob- mad, P. Mironowicz, M. Grünfeld, lem of Hidden Variables in Quantum Me- M. Bourennane, and P. Horodecki. Steering chanics. Indiana University Mathemat- is an essential feature of non-locality in ics Journal 17, 59–87 (1967). DOI: quantum theory. Nature Communications 10.1512/iumj.1968.17.17004. 9, 4244 (2018). DOI: 10.1038/s41467-018- 06255-5. [109] A. Cabello, J. Estebaranz, and G. García- Alcaine. Bell-Kochen-Specker theorem: A [119] S. Popescu and D. Rohrlich. Quantum proof with 18 vectors. Physics Letters A nonlocality as an axiom. Foundations 212, 183–187 (1996). DOI: 10.1016/0375- of Physics 24, 379–385 (1994). DOI: 9601(96)00134-x. 10.1007/bf02058098.

[110] A. Cabello. Experimentally Testable State- [120] S. Popescu. Nonlocality beyond quantum Independent Quantum Contextuality. Physi- mechanics. Nature Physics 10, 264–270 cal Review Letters 101, 210401 (2008). DOI: (2014). DOI: 10.1038/nphys2916. 10.1103/physrevlett.101.210401. [121] M. Pawłowski, T. Paterek, D. Kaszlikowski, [111] B. Marques, J. Ahrens, M. Nawareg, A. Ca- V. Scarani, A. Winter, and M. Żukowski. bello, and M. Bourennane. Experimen- Information causality as a physical princi- tal Observation of Hardy-Like Quantum ple. Nature 461, 1101–1104 (2009). DOI: Contextuality. Physical Review Letters 10.1038/nature08400. 113, 250403 (2014). DOI: 10.1103/physrevlett.113.250403. [122] M. Piani, M. Horodecki, P. Horodecki, and R. Horodecki. Properties of quan- [112] A. Grudka, K. Horodecki, M. Horodecki, tum nonsignaling boxes. Physical Review P. Horodecki, R. Horodecki, P. Joshi, A 74, 012305 (2006). DOI: 10.1103/phys- W. Kłobus, and A. Wójcik. Quantify- reva.74.012305. ing Contextuality. Physical Review Letters 112, 120401 (2014). DOI: 10.1103/phys- [123] K. Horodecki, A. Grudka, P. Joshi, revlett.112.120401. W. Kłobus, and J. Łodyga. Axiomatic approach to contextuality and nonlocality. [113] A. A. Klyachko, M. A. Can, S. Binicioğlu, Physical Review A 92, 032104 (2015). DOI: and A. S. Shumovsky. Simple Test for Hid- 10.1103/physreva.92.032104. den Variables in Spin-1 Systems. Physical Review Letters 101, 020403 (2008). DOI: [124] J.-D. Bancal and N. Gisin. Non-Local 10.1103/physrevlett.101.020403. Boxes for Networks. arXiv e-prints , arXiv:2102.03597v1 (2021). [114] N. D. Mermin. Simple uniﬁed form for the major no-hidden-variables theorems. Phys- [125] W. van Dam. Nonlocality & Communication ical Review Letters 65, 3373–3376 (1990). Complexity. Ph.D. thesis, Oxford (2000). DOI: 10.1103/physrevlett.65.3373. [126] G. Brassard, H. Buhrman, N. Linden, A. A. [115] I. Šupić and J. Bowles. Self-testing of quan- Méthot, A. Tapp, and F. Unger. Limit tum systems: a review. Quantum 4, 337 on Nonlocality in Any World in Which (2020). DOI: 10.22331/q-2020-09-30-337. Communication Complexity Is Not Trivial. Physical Review Letters 96, 250401 (2006). [116] S. Pironio, A. Acín, S. Massar, A. B. de la DOI: 10.1103/physrevlett.96.250401. Giroday, D. N. Matsukevich, P. Maunz, S. Olmschenk, D. Hayes, et al. Random [127] J. Grunhaus, S. Popescu, and D. Rohrlich. numbers certiﬁed by Bell’s theorem. Nature Jamming nonlocal quantum correlations. 464, 1021–1024 (2010). DOI: 10.1038/na- Physical Review A 53, 3781–3784 (1996). ture09008. DOI: 10.1103/physreva.53.3781.

23 [128] P. Horodecki and R. Ramanathan. The [138] K. Mattle, H. Weinfurter, P. G. Kwiat, and relativistic causality versus no-signaling A. Zeilinger. Dense Coding in Experimen- paradigm for multi-party correlations. Na- tal Quantum Communication. Physical Re- ture Communications 10, 1701 (2019). DOI: view Letters 76, 4656–4659 (1996). DOI: 10.1038/s41467-019-09505-2. 10.1103/physrevlett.76.4656.

[129] M. Eckstein, P. Horodecki, T. Miller, and [139] T. Schaetz, M. D. Barrett, D. Leibfried, R. Horodecki. Operational causality in J. Chiaverini, J. Britton, W. M. Itano, spacetime. Physical Review A 101, 042128 J. D. Jost, C. Langer, et al. Quantum (2020). DOI: 10.1103/physreva.101.042128. Dense Coding with Atomic Qubits. Physi- cal Review Letters 93, 040505 (2004). DOI: [130] T. Miller, M. Eckstein, P. Horodecki, and 10.1103/physrevlett.93.040505. R. Horodecki. Generally covariant N- [140] D. Wei, X. Yang, J. Luo, X. Sun, X. Zeng, particle dynamics. Journal of Geometry and M. Liu. NMR experimental implemen- and Physics 160, 103990 (2021). DOI: tation of three-parties quantum superdense 10.1016/j.geomphys.2020.103990. coding. Chinese Science Bulletin 49, 423– [131] A. Dragan and A. Ekert. Quantum prin- 426 (2004). DOI: 10.1007/bf02900957. ciple of relativity. New Journal of Physics [141] R. Liss and T. Mor. From Practice to 22 , 033038 (2020). DOI: 10.1088/1367- Theory: The “Bright Illumination” Attack 2630/ab76f7. on Quantum Key Distribution Systems. In Theory and Practice of Natural Comput- [132] M. Koashi and A. Winter. Monogamy of ing, vol.12494, 82–94. Springer International quantum entanglement and other correla- Publishing (2020). DOI: 10.1007/978-3-030- 69 tions. Physical Review A , 022309 (2004). 63000-3_7. DOI: 10.1103/physreva.69.022309. [142] L. Vaidman. Teleportation of quantum [133] M. P. Peloso, I. Gerhardt, C. Ho, A. Lamas- states. Physical Review A 49, 1473–1476 Linares, and C. Kurtsiefer. Daylight op- (1994). DOI: 10.1103/physreva.49.1473. eration of a free space, entanglement-based quantum key distribution system. New Jour- [143] S. L. Braunstein and H. J. Kimble. Telepor- nal of Physics 11, 045007 (2009). DOI: tation of Continuous Quantum Variables. 10.1088/1367-2630/11/4/045007. Physical Review Letters 80, 869–872 (1998). DOI: 10.1103/physrevlett.80.869. [134] M. Fujiwara, K. ichiro Yoshino, Y. Nambu, T. Yamashita, S. Miki, H. Terai, Z. Wang, [144] I. Marcikic, H. de Riedmatten, W. Tittel, M. Toyoshima, et al. Modiﬁed E91 proto- H. Zbinden, and N. Gisin. Long-distance col demonstration with hybrid entanglement teleportation of qubits at telecommunication 421 photon source. Optics Express 22, 13616 wavelengths. Nature , 509–513 (2003). (2014). DOI: 10.1364/oe.22.013616. DOI: 10.1038/nature01376. [145] R. Ursin, T. Jennewein, M. Aspelmeyer, [135] H. Y. Lim, T. Vergoossen, R. Beding- R. Kaltenbaek, M. Lindenthal, P. Walther, ton, X. Bai, A. Villar, A. Lohrmann, and and A. Zeilinger. Quantum teleportation other. Thermo-mechanical design for a across the Danube. Nature 430, 849–849 miniaturized quantum light source on board (2004). DOI: 10.1038/430849a. the SpooQy-1 CubeSat. arXiv e-prints , arXiv:2006.14442v1 (2020). [146] J. Yin, J.-G. Ren, H. Lu, Y. Cao, H.- L. Yong, Y.-P. Wu, C. Liu, S.-K. Liao, [136] C. C.-W. Lim, F. Xu, J.-W. Pan, and A. Ek- et al. Quantum teleportation and en- ert. Security analysis of quantum key dis- tanglement distribution over 100-kilometre tribution with small block length and its ap- free-space channels. Nature 488, 185–188 plication to quantum space communications. (2012). DOI: 10.1038/nature11332. arXiv e-prints, arXiv:2009.04882 (2020). [147] D. Gottesman and I. L. Chuang. Demon- [137] J. Barrett, L. Hardy, and A. Kent. No strating the viability of universal quantum Signaling and Quantum Key Distribution. computation using teleportation and single- Physical Review Letters 95, 010503 (2005). qubit operations. Nature 402, 390–393 DOI: 10.1103/physrevlett.95.010503. (1999). DOI: 10.1038/46503.

24 [148] X.-S. Ma, T. Herbst, T. Scheidl, D. Wang, Review Letters 101, 240501 (2008). DOI: S. Kropatschek, W. Naylor, B. Wittmann, 10.1103/physrevlett.101.240501. A. Mech, et al. Quantum teleportation over 143 kilometres using active feed- [158] S. Ishizaka and T. Hiroshima. Quantum forward. Nature 489, 269–273 (2012). DOI: teleportation scheme by selecting one of 10.1038/nature11472. multiple output ports. Physical Review A 79, 042306 (2009). DOI: 10.1103/phys- [149] C. Nölleke, A. Neuzner, A. Reiserer, reva.79.042306. C. Hahn, G. Rempe, and S. Ritter. Effi- cient Teleportation Between Remote Single- [159] P. Kopszak, M. Mozrzymas, M. Studzinski, Atom Quantum Memories. Physical Re- and M. Horodecki. Multiport based tele- view Letters 110, 140403 (2013). DOI: portation – transmission of a large amount 10.1103/physrevlett.110.140403. of quantum information. arXiv e-prints , arXiv:2008.00856v2 (2021). [150] W. Pfaff, B. J. Hensen, H. Bernien, S. B. van Dam, M. S. Blok, T. H. Taminiau, M. J. [160] S. Roy, T. Das, D. Das, A. Sen(De), and Tiggelman, R. N. Schouten, et al. Uncondi- U. Sen. How efficient is transport of quan- tional quantum teleportation between distant tum cargo through multiple highways? An- solid-state quantum bits. Science 345, 532– nals of Physics 422, 168281 (2020). DOI: 535 (2014). DOI: 10.1126/science.1253512. 10.1016/j.aop.2020.168281. [151] Y.-H. Luo, H.-S. Zhong, M. Erhard, X.- [161] M. Murao, D. Jonathan, M. B. Plenio, and L. Wang, L.-C. Peng, M. Krenn, X. Jiang, V. Vedral. Quantum telecloning and multi- L. Li, et al. Quantum Teleportation in particle entanglement. Physical Review A High Dimensions. Physical Review Letters 59, 156–161 (1999). DOI: 10.1103/phys- 123, 070505 (2019). DOI: 10.1103/physreva.59.156. revlett.123.070505. [162] S. Bose, V. Vedral, and P. L. Knight. [152] X.-M. Hu, C. Zhang, B.-H. Liu, Y. Cai, X.- Multiparticle generalization of entanglement J. Ye, Y. Guo, W.-B. Xing, C.-X. Huang, swapping. Physical Review A 57, 822–829 et al. Experimental High-Dimensional (1998). DOI: 10.1103/physreva.57.822. Quantum Teleportation. Physical Re- [163] J.-W. Pan, Z.-B. Chen, C.-Y. Lu, H. We- view Letters 125, 230501 (2020). DOI: infurter, A. Zeilinger, and M. Żukowski. 10.1103/physrevlett.125.230501. Multiphoton entanglement and interferome- [153] J.-G. Ren, P. Xu, H.-L. Yong, L. Zhang, try. Reviews of Modern Physics 84, 777–838 S.-K. Liao, J. Yin, W.-Y. Liu, W.-Q. Cai, (2012). DOI: 10.1103/revmodphys.84.777. et al. Ground-to-satellite quantum telepor- [164] J.-W. Pan, D. Bouwmeester, H. Weinfurter, tation. Nature 549, 70–73 (2017). DOI: and Z. A. Experimental entanglement swap- 10.1038/nature23675. ping: Entangling photons that never in- [154] N. Gisin. Entanglement 25 Years after teracted. Phys. Rev. Lett. 80, 3891–3894 Quantum Teleportation: Testing Joint Mea- (1998). surements in Quantum Networks. Entropy [165] R.-B. Jin, M. Takeoka, U. Takagi, 21, 325 (2019). DOI: 10.3390/e21030325. R. Shimizu, and M. Sasaki. Highly efficient [155] H. Salih, J. R. Hance, W. McCutcheon, entanglement swapping and teleportation at T. Rudolph, and J. Rarity. Determinis- telecom wavelength. Scientific Reports 5, tic Teleportation and Universal Computa- 9333 (2015). DOI: 10.1038/srep09333. tion Without Particle Exchange. arXiv e- [166] F. B. Basset, M. B. Rota, C. Schimpf, prints, arXiv:2009.05564 (2020). D. Tedeschi, K. D. Zeuner, S. F. C. da Silva, [156] M. Horodecki, P. Horodecki, and M. Reindl, V. Zwiller, et al. Entanglement R. Horodecki. General teleportation Swapping with Photons Generated on De- channel, singlet fraction, and quasidistil- mand by a Quantum Dot. Phys. Rev. Lett. lation. Physical Review A 60, 1888–1898 123, 160501 (Published 14 October 2019). (1999). DOI: 10.1103/physreva.60.1888. DOI: 10.1103/PhysRevLett.123.160501. [157] S. Ishizaka and T. Hiroshima. Asymptotic [167] Y. Zhang, M. Agnew, T. Roger, F. S. Teleportation Scheme as a Universal Pro- Roux, T. Konrad, D. Faccio, J. Leach, grammable Quantum Processor. Physical and A. Forbes. Simultaneous entanglement

25 swapping of multiple orbital angular mo- 77, 1413–1415 (1996). DOI: 10.1103/phys- mentum states of light. Nature Communica- revlett.77.1413. tions 8, 632 (2017). DOI: 10.1038/s41467- 017-00706-1. [177] S. L. Woronowicz. Nonextendible positive maps. Communications in Mathemat- [168] V. Zapatero and M. Curty. Long-distance ical Physics 51, 243–282 (1976). DOI: device-independent quantum key distribu- 10.1007/bf01617922. tion. Scientific Reports 9, 17749 (2019). [178] M. Horodecki, P. Horodecki, and DOI: 10.1038/s41598-019-53803-0. R. Horodecki. Separability of mixed [169] H.-J. Briegel, W. Dür, J. I. Cirac, and states: necessary and sufficient conditions. P. Zoller. Quantum Repeaters: The Role Physics Letters A 223, 1–8 (1996). DOI: of Imperfect Local Operations in Quantum 10.1016/s0375-9601(96)00706-2. Communication. Physical Review Letters [179] B. M. Terhal. Bell inequalities and the 81, 5932–5935 (1998). DOI: 10.1103/phys- separability criterion. Physics Letters A revlett.81.5932. 271, 319–326 (2000). DOI: 10.1016/s0375- [170] L.-M. Duan, M. D. Lukin, J. I. Cirac, and 9601(00)00401-1. P. Zoller. Long-distance quantum commu- [180] M. Lewenstein, B. Kraus, J. I. Cirac, and nication with atomic ensembles and linear P. Horodecki. Optimization of entanglement optics. Nature 414, 413–418 (2001). DOI: witnesses. Physical Review A 62, 052310 10.1038/35106500. (2000). DOI: 10.1103/physreva.62.052310. [171] B. K. Behera, S. Seth, A. Das, and P. K. [181] M. Lewenstein, B. Kraus, P. Horodecki, and Panigrahi. Demonstration of entanglement J. I. Cirac. Characterization of separable purification and swapping protocol to design states and entanglement witnesses. Phys- quantum repeater in IBM quantum com- ical Review A 63, 044304 (2001). DOI: 18 puter. Quantum Information Processing . 10.1103/physreva.63.044304. DOI: 10.1007/s11128-019-2229-2. [182] A. Jamiołkowski. Linear transformations [172] M. Zopf, R. Keil, Y. Chen, J. Yang, which preserve trace and positive semidef- D. Chen, F. Ding, and O. G. Schmidt. En- initeness of operators. Reports on Mathe- tanglement Swapping with Semiconductor- matical Physics 3, 275–278 (1972). DOI: generated Photons. Phys. Rev. Lett. 10.1016/0034-4877(72)90011-0. 123, 160502 (2019). DOI: 10.1103/Phys- RevLett.123.160502. [183] M. Barbieri, F. D. Martini, G. D. Nepi, P. Mataloni, G. M. D’Ariano, and C. Mac- [173] Z. Zhang and X. Man. Multiparty Quan- chiavello. Detection of Entanglement with tum Secret Sharing Based on Entangle- Polarized Photons: Experimental Realiza- ment Swapping. Physical Review A 72, tion of an Entanglement Witness. Physi- 022303 (June 2004). DOI: 10.1103/Phys- cal Review Letters 91. DOI: 10.1103/phys- RevA.72.022303. revlett.91.227901. [174] N. T. V. Luu and S. Shimamoto. Ad- [184] M. Bourennane, M. Eibl, C. Kurtsiefer, vanced Multiparty Quantum Secret Shar- S. Gaertner, H. Weinfurter, O. Gühne, ing using Entanglement Swapping. In P. Hyllus, D. Bruß, et al. Experimental De- 2006 2nd International Conference on tection of Multipartite Entanglement using Information Communication Technologies, Witness Operators. Physical Review Let- volume 2, 2051–2056 (2006). DOI: ters 92, 087902 (2004). DOI: 10.1103/phys- 10.1109/ICTTA.2006.1684717. revlett.92.087902. [175] L. Gurvits. Classical deterministic complex- [185] C. F. Roos. Control and Measurement ity of Edmonds' Problem and quantum en- of Three-Qubit Entangled States. Science tanglement. In Proceedings of the thirty-fifth 304, 1478–1480 (2004). DOI: 10.1126/sci- ACM symposium on Theory of computing - ence.1097522. STOC '03, 10. ACM Press (2003). DOI: 10.1145/780542.780545. [186] J. B. Altepeter, E. R. Jeffrey, P. G. Kwiat, S. Tanzilli, N. Gisin, and A. Acín. [176] A. Peres. Separability Criterion for Den- Experimental Methods for Detecting En- sity Matrices. Physical Review Letters tanglement. Physical Review Letters

26 95, 033601 (2005). DOI: 10.1103/phys- Physical Review A 84, 062305 (2011). DOI: revlett.95.033601. 10.1103/physreva.84.062305.

[187] H. Häffner, W. Hänsel, C. F. Roos, J. Ben- [196] M. Markiewicz, A. Kołodziejski, Z. Puchała, helm, D. C. al kar, M. Chwalla, T. Körber, A. Rutkowski, T. Tylec, and W. Laskowski. U. D. Rapol, et al. Scalable multiparticle Unified approach to geometric and positive- entanglement of trapped ions. Nature 438, map-based nonlinear entanglement identi- 643–646 (2005). DOI: 10.1038/nature04279. fiers. Physical Review A 97, 042339 (2018). DOI: 10.1103/physreva.97.042339. [188] H. Mikami, Y. Li, K. Fukuoka, and [197] C.-J. Zhang, Y.-S. Zhang, S. Zhang, and G.- T. Kobayashi. New High-Efficiency Source C. Guo. Entanglement detection beyond the of a Three-PhotonWState and its Full computable cross-norm or realignment cri- Characterization Using Quantum State To- terion. Physical Review A 77, 060301(R) mography. Physical Review Letters 95, (2008). DOI: 10.1103/physreva.77.060301. 150404 (2005). DOI: 10.1103/physrevlett.95.150404. [198] G. Sarbicki, G. Scala, and D. Chruściński. Family of multipartite separability crite- [189] N. K. Langford, T. J. Weinhold, ria based on a correlation tensor. Physi- R. Prevedel, K. J. Resch, A. Gilchrist, J. L. cal Review A 101, 012341 (2020). DOI: O’Brien, G. J. Pryde, and A. G. White. 10.1103/physreva.101.012341. Demonstration of a Simple Entangling Opti- cal Gate and Its Use in Bell-State Analysis. [199] G. Sarbicki, G. Scala, and D. Chruś- Physical Review Letters 95, 210504 (2005). ciński. Detection power of separability cri- DOI: 10.1103/physrevlett.95.210504. teria based on a correlation tensor: a case study. arXiv e-prints , arXiv:2011.10159 [190] W. Laskowski, D. Richart, C. Schwemmer, (2020). T. Paterek, and H. Weinfurter. Experimen- tal Schmidt Decomposition and State Inde- [200] P. Horodecki. Separability criterion and pendent Entanglement Detection. Physical inseparable mixed states with positive par- 232 Review Letters 108. DOI: 10.1103/phys- tial transposition. Physics Letters A , revlett.108.240501. 333–339 (1997). DOI: 10.1016/s0375- 9601(97)00416-7. [191] B. Dirkse, M. Pompili, R. Hanson, M. Wal- [201] C. H. Bennett, G. Brassard, S. Popescu, ter, and S. Wehner. Witnessing entangle- B. Schumacher, J. A. Smolin, and W. K. ment in experiments with correlated noise. Wootters. Purification of Noisy En- 5 Quantum Science and Technology , 035007 tanglement and Faithful Teleportation via (2020). DOI: 10.1088/2058-9565/ab8d88. Noisy Channels. Physical Review Letters 76 [192] C. Branciard, D. Rosset, Y.-C. Liang, and , 722–725 (1996). DOI: 10.1103/phys- N. Gisin. Measurement-Device-Independent revlett.76.722. Entanglement Witnesses for All Entangled [202] D. Deutsch, A. Ekert, R. Jozsa, C. Macchi- Quantum States. Physical Review Letters avello, S. Popescu, and A. Sanpera. Quan- 110, 060405 (2013). DOI: 10.1103/phys- tum Privacy Amplification and the Security revlett.110.060405. of Quantum Cryptography over Noisy Chan- nels. Physical Review Letters 77, 2818–2821 [193] M. Oszmaniec and M. Kuś. Univer- (1996). DOI: 10.1103/physrevlett.77.2818. sal framework for entanglement detection. Physical Review A 88, 052328 (2013). DOI: [203] M. Horodecki, P. Horodecki, and 10.1103/physreva.88.052328. R. Horodecki. Inseparable Two Spin- 12Density Matrices Can Be Distilled to [194] P. Badziag, Č. Brukner, W. Laskowski, a Singlet Form. Physical Review Letters T. Paterek, and M. Żukowski. Experi- 78, 574–577 (1997). DOI: 10.1103/phys- mentally Friendly Geometrical Criteria for revlett.78.574. Entanglement. Physical Review Letters 100, 140403 (2008). DOI: 10.1103/phys- [204] M. Horodecki, P. Horodecki, and revlett.100.140403. R. Horodecki. Mixed-State Entangle- ment and Distillation: Is there a “Bound” [195] W. Laskowski, M. Markiewicz, T. Paterek, Entanglement in Nature? Physical Re- and M. Żukowski. Correlation-tensor cri- view Letters 80, 5239–5242 (1998). DOI: teria for genuine multiqubit entanglement. 10.1103/physrevlett.80.5239.

27 [205] R. Horodecki. Bound entanglement – [215] K. Horodecki, M. Horodecki, P. Horodecki, mysterious invention of nature. Euro- and J. Oppenheim. General Paradigm physics News 41, 21–24 (2010). DOI: for Distilling Classical Key From Quantum 10.1051/epn/2010603. States. IEEE Transactions on Informa- tion Theory 55, 1898–1929 (2009). DOI: [206] E. Amselem and M. Bourennane. Ex- 10.1109/tit.2008.2009798. perimental four-qubit bound entanglement. Nature Physics 5, 748–752 (2009). DOI: [216] K. Dobek, M. Karpiński, R. Demkowicz- 10.1038/nphys1372. Dobrzański, K. Banaszek, and P. Horodecki. Experimental Extraction of Secure Correla- [207] J. Lavoie, R. Kaltenbaek, M. Piani, and tions from a Noisy Private State. Physical K. J. Resch. Experimental Bound Entan- Review Letters 106, 030501 (2011). DOI: glement in a Four-Photon State. Physical 10.1103/physrevlett.106.030501. Review Letters 105, 130501 (2010). DOI: 10.1103/physrevlett.105.130501. [217] T. Vértesi and N. Brunner. Disproving the Peres conjecture by showing Bell non- [208] J. T. Barreiro, P. Schindler, O. Gühne, locality from bound entanglement. Nature T. Monz, M. Chwalla, C. F. Roos, M. Hen- Communications 5, 5297 (2014). DOI: nrich, and R. Blatt. Experimental multi- 10.1038/ncomms6297. particle entanglement dynamics induced by decoherence. Nature Physics 6, 943–946 [218] Ł. Czekaj, A. Przysiężna, M. Horodecki, and (2010). DOI: 10.1038/nphys1781. P. Horodecki. Quantum metrology: Heisen- berg limit with bound entanglement. Phys- [209] H. Kampermann, D. Bruß, X. Peng, and ical Review A 92, 062303 (2015). DOI: D. Suter. Experimental generation of 10.1103/physreva.92.062303. pseudo-bound-entanglement. Physical Re- view A 81, 040304(R) (2010). DOI: [219] G. Tóth and T. Vértesi. Quantum States 10.1103/physreva.81.040304. with a Positive Partial Transpose are Use- ful for Metrology. Physical Review Letters [210] J. DiGuglielmo, A. Samblowski, B. Hage, 120, 020506 (2018). DOI: 10.1103/phys- C. Pineda, J. Eisert, and R. Schnabel. Ex- revlett.120.020506. perimental Unconditional Preparation and Detection of a Continuous Bound Entan- [220] K. F. Pál, G. Tóth, E. Bene, and T. Vértesi. gled State of Light. Physical Review Letters Bound entangled “singlets” for quantum 107, 240503 (2011). DOI: 10.1103/phys- metrology. arXiv e-prints, arXiv:2002.12409 revlett.107.240503. (2020). [221] R. Arnon-Friedman, F. Dupuis, O. Fawzi, [211] G. Tóth, C. Knapp, O. Gühne, and H. J. R. Renner, and T. Vidick. Practical device- Briegel. Optimal Spin Squeezing Inequalities independent quantum cryptography via en- Detect Bound Entanglement in Spin Models. tropy accumulation. Nature Communica- Physical Review Letters 99, 250405 (2007). tions 9. DOI: 10.1038/s41467-017-02307-4. DOI: 10.1103/physrevlett.99.250405. [222] R. Arnon-Friedman and F. Leditzky. Upper [212] A. Ferraro, D. Cavalcanti, A. García-Saez, bounds on device-independent quantum key and A. Acín. Thermal Bound Entanglement distribution rates and a revised Peres con- in Macroscopic Systems and Area Law. jecture. arXiv e-prints , arXiv:2005.12325 Physical Review Letters 100, 080502 (2008). (2020). DOI: 10.1103/physrevlett.100.080502. [223] M. Christandl, R. Ferrara, and [213] P. Horodecki, M. Horodecki, and K. Horodecki. Upper bounds on the rate in R. Horodecki. Bound Entanglement device-independent quantum key distribu- Can Be Activated. Physical Review Letters tion. arXiv e-prints , arXiv:2005.13511v2 82, 1056–1059 (1999). DOI: 10.1103/phys- (2020). revlett.82.1056. [224] E. Schrödinger. Die gegenwärtige Situa- [214] K. Horodecki, M. Horodecki, P. Horodecki, tion in der Quantenmechanik. Naturwis- and J. Oppenheim. Secure Key from Bound senschaften 23, 807-812 (1935). Entanglement. Physical Review Letters 94, 160502 (2005). DOI: 10.1103/phys- [225] R. Horodecki and P. Horodecki. Quan- revlett.94.160502. tum redundancies and local realism. Physics

28 Letters A 194, 147–152 (1994). DOI: [236] I. Devetak and A. Winter. Distillation of 10.1016/0375-9601(94)91275-0. secret key and entanglement from quantum states. Proceedings of the Royal Society [226] N. J. Cerf and C. Adami. Negative En- A: Mathematical, Physical and Engineer- tropy and Information in Quantum Mechan- ing Sciences 461, 207–235 (2005). DOI: ics. Physical Review Letters 79, 5194–5197 10.1098/rspa.2004.1372. (1997). DOI: 10.1103/physrevlett.79.5194. [237] R. Horodecki and M. Horodecki. [227] B. Schumacher and M. D. Westmore- Information-theoretic aspects of insepa- land. Sending classical information via rability of mixed states. Physical Review A noisy quantum channels. Physical Review 54, 1838–1843 (1996). DOI: 10.1103/phys- 56 A , 131–138 (1997). DOI: 10.1103/physreva.54.1838. reva.56.131. [238] P. Horodecki and A. Ekert. Method for Di- [228] A. Holevo. The capacity of the quantum rect Detection of Quantum Entanglement. channel with general signal states. IEEE Physical Review Letters 89, 125003 (2002). 44 Transactions on Information Theory , DOI: 10.1103/physrevlett.89.127902. 269–273 (1998). DOI: 10.1109/18.651037. [239] P. Horodecki. Measuring Quantum Entan- [229] I. Devetak. The Private Classical Capac- glement without Prior State Reconstruction. ity and Quantum Capacity of a Quantum Physical Review Letters 90, 167901 (2003). Channel. IEEE Transactions on Infor- DOI: 10.1103/physrevlett.90.167901. mation Theory 51, 44–55 (2005). DOI: 10.1109/tit.2004.839515. [240] F. A. Bovino, G. Castagnoli, A. Ekert, P. Horodecki, C. M. Alves, and A. V. [230] C. H. Bennett, I. Devetak, P. W. Shor, Sergienko. Direct Measurement of Nonlin- and J. A. Smolin. Inequalities and Separa- ear Properties of Bipartite Quantum States. tions Among Assisted Capacities of Quan- Physical Review Letters 95, 240407 (2005). tum Channels. Physical Review Letters DOI: 10.1103/physrevlett.95.240407. 96, 150502 (2006). DOI: 10.1103/physrevlett.96.150502. [241] R. Islam, R. Ma, P. M. Preiss, M. E. Tai, A. Lukin, M. Rispoli, and M. Greiner. [231] K. Horodecki, M. Horodecki, P. Horodecki, Measuring entanglement entropy in a quan- D. Leung, and J. Oppenheim. Quan- tum many-body system. Nature 528, 77–83 tum Key Distribution Based on Private (2015). DOI: 10.1038/nature15750. States: Unconditional Security Over Un- trusted Channels With Zero Quantum Ca- [242] A. M. Kaufman, M. E. Tai, A. Lukin, pacity. IEEE Transactions on Informa- M. Rispoli, R. Schittko, P. M. Preiss, tion Theory 54, 2604–2620 (2008). DOI: and M. Greiner. Quantum thermalization 10.1109/tit.2008.921870. through entanglement in an isolated many- body system. Science 353, 794–800 (2016). [232] K. Li, A. Winter, X. Zou, and G. Guo. DOI: 10.1126/science.aaf6725. Private Capacity of Quantum Channels is Not Additive. Physical Review Letters [243] N. M. Linke, S. Johri, C. Figgatt, K. A. 103, 120501 (2009). DOI: 10.1103/phys- Landsman, A. Y. Matsuura, and C. Mon- revlett.103.120501. roe. Measuring the Rényi entropy of a two-site Fermi-Hubbard model on a trapped [233] S. Bäuml, M. Christandl, K. Horodecki, and ion quantum computer. Physical Review A. Winter. Limitations on quantum key re- A 98, 052334 (2018). DOI: 10.1103/phys- peaters. Nature Communications 6. DOI: reva.98.052334. 10.1038/ncomms7908. [244] C. M. Alves and D. Jaksch. Multipartite [234] D. Slepian and J. Wolf. Noiseless cod- Entanglement Detection in Bosons. Physi- ing of correlated information sources. cal Review Letters 93, 110501 (2004). DOI: IEEE Transactions on Information 10.1103/physrevlett.93.110501. Theory 19, 471–480 (1973). DOI: [245] A. J. Daley, H. Pichler, J. Schachenmayer, 10.1109/tit.1973.1055037. and P. Zoller. Measuring Entanglement [235] M. Horodecki, J. Oppenheim, and A. Win- Growth in Quench Dynamics of Bosons in ter. Partial quantum information. Nature an Optical Lattice. Physical Review Letters 436, 673–676 (2005). DOI: 10.1038/na- 109, 020505 (2012). DOI: 10.1103/phys- ture03909. revlett.109.020505.

29 [246] T. Brydges, A. Elben, P. Jurcevic, B. Ver- T. Sleator, J. A. Smolin, et al. Elemen- mersch, C. Maier, B. P. Lanyon, P. Zoller, tary gates for quantum computation. Phys- R. Blatt, et al. Probing Rényi entan- ical Review A 52, 3457–3467 (1995). DOI: glement entropy via randomized measure- 10.1103/physreva.52.3457. ments. Science 364, 260–263 (2019). DOI: 10.1126/science.aau4963. [255] D. P. DiVincenzo. The Physical Im- plementation of Quantum Compu- [247] S. J. van Enk and C. W. J. Beenakker. tation. Fortschritte der Physik 48, Measuring Tr ρn on Single Copies of ρ Us- 771–783 (2000). DOI: 10.1002/1521- ing Random Measurements. Physical Re- 3978(200009)48:9/11<771::aid- view Letters 108, 110503 (2012). DOI: prop771>3.0.co;2-e. 10.1103/physrevlett.108.110503. [256] L. M. K. Vandersypen, M. Steﬀen, [248] A. Elben, B. Vermersch, M. Dalmonte, G. Breyta, C. S. Yannoni, M. H. Sherwood, J. Cirac, and P. Zoller. Rényi Entropies and I. L. Chuang. Experimental realization from Random Quenches in Atomic Hubbard of Shor's quantum factoring algorithm using 414 and Spin Models. Physical Review Letters nuclear magnetic resonance. Nature , 120, 050406 (2018). DOI: 10.1103/phys- 883–887 (2001). DOI: 10.1038/414883a. revlett.120.050406. [257] W.-L. Yang, H. Wei, C.-Y. Chen, and M. Feng. Implementation of a many- [249] A. Elben, B. Vermersch, C. F. Roos, and qubit Grover search with trapped ultra- P. Zoller. Statistical correlations between cold ions. Journal of the Optical Soci- locally randomized measurements: A tool- ety of America B 25, 1720 (2008). DOI: box for probing entanglement in many- 10.1364/josab.25.001720. body quantum states. Physical Review A 99, 052323 (2019). DOI: 10.1103/phys- [258] P. Bianucci, A. Muller, C. K. Shih, Q. Q. reva.99.052323. Wang, Q. K. Xue, and C. Piermarocchi. Experimental realization of the one qubit [250] H.-Y. Huang, R. Kueng, and J. Preskill. Deutsch-Jozsa algorithm in a quantum dot. Predicting many properties of a quantum Physical Review B 69, 161303(R) (2004). system from very few measurements. Na- DOI: 10.1103/physrevb.69.161303. ture Physics 16, 1050–1057 (2020). DOI: 10.1038/s41567-020-0932-7. [259] S. Lloyd. Universal Quantum Simulators. Science 273, 1073–1078 (1996). DOI: [251] V. Trávníček, K. Bartkiewicz, A. Černoch, 10.1126/science.273.5278.1073. and K. Lemr. Experimental measurement of a nonlinear entanglement witness by hyper- [260] A. Peruzzo, J. McClean, P. Shadbolt, M.-H. entangling two-qubit states. Physical Review Yung, X.-Q. Zhou, P. J. Love, A. Aspuru- A 98, 032307 (2018). DOI: 10.1103/phys- Guzik, and J. L. O’Brien. A variational reva.98.032307. eigenvalue solver on a photonic quantum processor. Nature Communications 5, 4213 [252] K. Bartkiewicz, K. Lemr, A. Černoch, (2014). DOI: 10.1038/ncomms5213. and A. Miranowicz. Bell nonlocality [261] Y. Alexeev, D. Bacon, K. R. Brown, and fully entangled fraction measured in R. Calderbank, L. D. Carr, F. T. Chong, an entanglement-swapping device without B. DeMarco, D. Englund, et al. Quantum quantum state tomography. Physical Re- Computer Systems for Scientiﬁc Discovery. 95 view A , 030102(R) (2017). DOI: arXiv e-prints, arXiv:1912.07577 (2019). 10.1103/physreva.95.030102. [262] A. M. Steane. Simple quantum error- [253] A. Elben, R. Kueng, H.-Y. R. Huang, correcting codes. Physical Review A 54, R. van Bijnen, C. Kokail, M. Dalmonte, 4741–4751 (1996). DOI: 10.1103/phys- P. Calabrese, B. Kraus, et al. Mixed- reva.54.4741. State Entanglement from Local Random- ized Measurements. Physical Review Letters [263] D. Aharonov and M. Ben-Or. Fault-tolerant 125, 200501 (2020). DOI: 10.1103/phys- quantum computation with constant error. revlett.125.200501. In Proceedings of the twenty-ninth annual ACM symposium on Theory of computing [254] A. Barenco, C. H. Bennett, R. Cleve, - STOC '97, 176. ACM Press (1997). DOI: D. P. DiVincenzo, N. Margolus, P. Shor, 10.1145/258533.258579.

30 [264] E. Knill, R. Laflamme, and W. H. Zurek. Physical Review A 85, 022321 (2012). DOI: Resilient Quantum Computation. Science 10.1103/physreva.85.022321. 279, 342–345 (1998). DOI: 10.1126/science.279.5349.342. [276] G. Tóth and I. Apellaniz. Quantum metrology from a quantum information science [265] J. I. Cirac and P. Zoller. Quantum Com- perspective. Journal of Physics A: Mathe- putations with Cold Trapped Ions. Physical matical and Theoretical 47, 424006 (2014). Review Letters 74, 4091–4094 (1995). DOI: DOI: 10.1088/1751-8113/47/42/424006. 10.1103/physrevlett.74.4091. [277] T. Xie et al. Beating the Standard [266] F. Arute, K. Arya, R. Babbush, D. Ba- Quantum Limit under Ambient Conditions con, J. C. Bardin, R. Barends, R. Biswas, with Solid-State Spins. arXiv e-prints , S. Boixo, et al. Quantum supremacy us- arXiv:2101.12048v1 (2021). ing a programmable superconducting processor. Nature 574, 505–510 (2019). DOI: [278] S. F. Huelga, C. Macchiavello, T. Pelliz- 10.1038/s41586-019-1666-5. zari, A. K. Ekert, M. B. Plenio, and J. I. Cirac. Improvement of Frequency Standards [267] S. Aaronson and A. Arkhipov. The Compu- with Quantum Entanglement. Physical Re- tational Complexity of Linear Optics. The- view Letters 79, 3865–3868 (1997). DOI: ory of Computing 9, 143–252 (2013). DOI: 10.1103/physrevlett.79.3865. 10.4086/toc.2013.v009a004. [279] R. Demkowicz-Dobrzański, U. Dorner, B. J. [268] H.-S. Zhong, H. Wang, Y.-H. Deng, M.-C. Smith, J. S. Lundeen, W. Wasilewski, Chen, L.-C. Peng, Y.-H. Luo, J. Qin, D. Wu, K. Banaszek, and I. A. Walmsley. Quantum et al. Quantum computational advantage us- phase estimation with lossy interferometers. ing photons. Science 370, 1460-1463 (2020). Physical Review A 80, 013825 (2009). DOI: [269] P. Ball. Physicists in China challenge 10.1103/physreva.80.013825. Google’s ‘quantum advantage’. Nature 588, [280] Y. Matsuzaki, S. C. Benjamin, and J. Fitzsi- 380–380 (2020). DOI: 10.1038/d41586-020- mons. Magnetic field sensing beyond the 03434-7. standard quantum limit under the effect of [270] C. M. Caves. Quantum-mechanical noise decoherence. Physical Review A 84, 012103 in an interferometer. Physical Review D (2011). DOI: 10.1103/physreva.84.012103. 23 , 1693–1708 (1981). DOI: 10.1103/phys- [281] R. Demkowicz-Dobrzański, J. Kołodyński, revd.23.1693. and M. Guţă. The elusive Heisenberg limit [271] V. Giovannetti. Quantum-Enhanced Mea- in quantum-enhanced metrology. Nature surements: Beating the Standard Quantum Communications 3, 1063 (2012). DOI: Limit. Science 306, 1330–1336 (2004). DOI: 10.1038/ncomms2067. 10.1126/science.1104149. [282] A. W. Chin, S. F. Huelga, and M. B. Ple- [272] V. Giovannetti, S. Lloyd, and L. Maccone. nio. Quantum Metrology in Non-Markovian Quantum Metrology. Physical Review Let- Environments. Physical Review Letters ters 96, 010401 (2006). DOI: 10.1103/phys- 109, 233601 (2012). DOI: 10.1103/physrevlett.96.010401. revlett.109.233601. [273] L. Pezzé and A. Smerzi. Entangle- [283] R. Demkowicz-Dobrzański, J. Czajkowski, ment, Nonlinear Dynamics, and the Heisen- and P. Sekatski. Adaptive Quantum Metrol- berg Limit. Physical Review Letters ogy under General Markovian Noise. Phys- 102, 100401 (2009). DOI: 10.1103/physical Review X 7, 041009 (2017). DOI: revlett.102.100401. 10.1103/physrevx.7.041009. [274] G. Tóth. Multipartite entanglement and [284] S. Zhou, M. Zhang, J. Preskill, and L. Jiang. high-precision metrology. Physical Review Achieving the Heisenberg limit in quantum A 85, 022322 (2012). DOI: 10.1103/phys- metrology using quantum error correction. reva.85.022322. Nature Communications 9, 78 (2018). DOI: 10.1038/s41467-017-02510-3. [275] P. Hyllus, W. Laskowski, R. Krischek, C. Schwemmer, W. Wieczorek, H. Wein- [285] K. Chabuda, J. Dziarmaga, T. J. Osborne, furter, L. Pezzé, and A. Smerzi. Fisher in- and R. Demkowicz-Dobrzański. Tensor- formation and multiparticle entanglement. network approach for quantum metrology

31 in many-body quantum systems. Nature [295] W. H. Zurek. Quantum Darwinism. Na- Communications 11, 250 (2020). DOI: ture Physics 5, 181–188 (2009). DOI: 10.1038/s41467-019-13735-9. 10.1038/nphys1202. [286] R. Demkowicz-Dobrzański, K. Banaszek, [296] F. G. S. L. Brandão, M. Piani, and and R. Schnabel. Fundamental quantum P. Horodecki. Generic emergence of clas- interferometry bound for the squeezed-light- sical features in quantum Darwinism. Na- enhanced gravitational wave detector GEO ture Communications 6, 7908 (2015). DOI: 600. Physical Review A 88, 041802(R) 10.1038/ncomms8908. (2013). DOI: 10.1103/physreva.88.041802. [297] T. P. Le and A. Olaya-Castro. Strong [287] A. Franzen, B. Hage, J. DiGuglielmo, J. Fi- Quantum Darwinism and Strong Indepen- urášek, and R. Schnabel. Experimental dence are Equivalent to Spectrum Broad- Demonstration of Continuous Variable Pu- cast Structure. Physical Review Letters rification of Squeezed States. Physical Re- 122, 010403 (2019). DOI: 10.1103/phys- view Letters 97, 150505 (2006). DOI: revlett.122.010403. 10.1103/physrevlett.97.150505. [298] J. Korbicz, P. Horodecki, and R. Horodecki. [288] M. Parniak, S. Borówka, K. Boroszko, Objectivity in a Noisy Photonic Environ- W. Wasilewski, K. Banaszek, and ment through Quantum State Information R. Demkowicz-Dobrzański. Beating Broadcasting. Physical Review Letters 112 the Rayleigh Limit Using Two-Photon , 120402 (2014). DOI: 10.1103/phys- Interference. Physical Review Letters revlett.112.120402. 121, 250503 (2018). DOI: 10.1103/phys- [299] R. Horodecki, J. K. Korbicz, and revlett.121.250503. P. Horodecki. Quantum origins of objectivity. Physical Review A 91, 032122 [289] P. Hyllus, O. Gühne, and A. Smerzi. Not (2015). DOI: 10.1103/physreva.91.032122. all pure entangled states are useful for sub- shot-noise interferometry. Physical Review [300] M.-C. Chen, H.-S. Zhong, Y. Li, D. Wu, X.- A 82. DOI: 10.1103/physreva.82.012337. L. Wang, L. Li, N.-L. Liu, C.-Y. Lu, et al. Emergence of classical objectivity of quan- [290] G. Tóth, T. Vértesi, P. Horodecki, and tum Darwinism in a photonic quantum sim- R. Horodecki. Activating Hidden Metro- ulator. Science Bulletin 64, 580–585 (2019). logical Usefulness. Physical Review Letters DOI: 10.1016/j.scib.2019.03.032. 125, 020402 (2020). DOI: 10.1103/physrevlett.125.020402. [301] C. M. Scandolo, R. Salazar, J. K. Korbicz, and P. Horodecki. The origin of objectivity [291] C. H. Bennett, A. Grudka, M. Horodecki, in all fundamental causal theories. arXiv e- P. Horodecki, and R. Horodecki. Postulates prints, arXiv:1805.12126v6 (2020). for measures of genuine multipartite correlations. Physical Review A 83, 012312 (2011). [302] R. Colbeck and R. Renner. Free randomness DOI: 10.1103/physreva.83.012312. can be amplified. Nature Physics 8, 450–453 (2012). DOI: 10.1038/nphys2300. [292] C. H. Bennett, D. P. DiVincenzo, C. A. Fuchs, T. Mor, E. Rains, P. W. Shor, J. A. [303] R. Gallego, L. Masanes, G. D. L. Torre, Smolin, and W. K. Wootters. Quantum C. Dhara, L. Aolita, and A. Acín. Full nonlocality without entanglement. Physi- randomness from arbitrarily deterministic 4 cal Review A 59, 1070–1091 (1999). DOI: events. Nature Communications , 2654 10.1103/physreva.59.1070. (2013). DOI: 10.1038/ncomms3654. [304] F. G. S. L. Brandão, R. Ramanathan, [293] G. Smith and J. Yard. Quantum Com- A. Grudka, K. Horodecki, M. Horodecki, munication with Zero-Capacity Channels. P. Horodecki, T. Szarek, and H. Wojewódka. Science 321, 1812–1815 (2008). DOI: Realistic noise-tolerant randomness amplifi- 10.1126/science.1162242. cation using finite number of devices. Na- [294] D. P. DiVincenzo, M. Horodecki, D. W. ture Communications 7, 11345 (2016). DOI: Leung, J. A. Smolin, and B. M. Ter- 10.1038/ncomms11345. hal. Locking Classical Correlations in [305] P. Horodecki, Łukasz Rudnicki, and Quantum States. Physical Review Letters K. Życzkowski. Five open problems in 92, 067902 (2004). DOI: 10.1103/phys- quantum information. arXiv e-prints , revlett.92.067902. arXiv:2002.03233v2 (2020).