DOCSLIB.ORG

Approximate Inference on Planar Graphs Using Loop Calculus and Belief Propagation

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Approximate Inference on Planar Graphs Using Loop Calculus and Belief Propagation Approximate inference on planar graphs using Loop Calculus and Belief Propagation Vicen¸cG´omez Michael Chertkov Hilbert J. Kappen Theoretical Division and Center for Nonlinear Studies Radboud University Nijmegen, Los Alamos National Laboratory Donders Institute for Brain, Cognition and Behaviour Los Alamos, NM 87545 6525 EZ Nijmegen, The Netherlands Abstract as hypothesis testing or parameter estimation. Ex- act computation of this quantity is only feasible when the graph is not too complex, or equivalently, when We introduce novel results for approximate its tree-width is small. Currently many methods are inference on planar graphical models using devoted to approximate this quantity. the loop calculus framework. The loop cal- culus (Chertkov and Chernyak, 2006b) allows The belief propagation (BP) algorithm (Pearl, 1988) is to express the exact partition function Z of at the core of many of these methods. Initially thought a graphical model as a finite sum of terms as an exact algorithm for tree graphs, it is widely used that can be evaluated once the belief prop- as an approximation method for loopy graphs (Mur- agation (BP) solution is known. In general, phy et al., 1999; Frey and MacKay, 1998). The exact full summation over all correction terms is partition function Z is explicitly related to the BP ap- intractable. We develop an algorithm for proximation through the loop calculus framework in- the approach presented in Chertkov et al. troduced by Chertkov and Chernyak (2006b). Loop (2008) which represents an efficient trunca- calculus allows to express Z as a finite sum of terms tion scheme on planar graphs and a new rep- (loop series) that can be evaluated once the BP so- resentation of the series in terms of Pfaffians lution is known. Each term maps uniquely to a sub- of matrices. We analyze in detail both the graph, also denoted as a generalized loop, where the loop series and the Pfaffian series for mod- connectivity of any node within the subgraph is at els with binary variables and pairwise in- least degree 2. Summation of the entire loop series teractions, and show that the first term of is a hard combinatorial task since the number of gen- the Pfaffian series can provide very accurate eralized loops is typically exponential in the size of the approximations. The algorithm outperforms graph. However, different approximations can be ob- previous truncation schemes of the loop series tained by considering different subsets of generalized and is competitive with other state-of-the-art loops in the graph. methods for approximate inference. Although it has been shown empirically (G´omez et al., 2007; Chertkov and Chernyak, 2006a) that truncating this series may provide efficient corrections to the ini- 1 Introduction tial BP approximation, a formal characterization of the classes of tractable problems via loop calculus still Graphical models are popular tools widely used in remains as an open question. The work of Chertkov many areas which require modeling of uncertainty. et al. (2008) represents a step in this direction, where They provide an effective approach through a com- it was shown that for any graphical model, summation pact representation of the joint probability distribu- of a certain subset of terms can be mapped to a sum- tion. The two most common types of graphical mod- mation of weighted perfect matchings on an extended els are Bayesian Networks (BN) and Markov Random graph. For planar graphs (graphs that can be embed- Fields (MRFs). ded into a plane without crossing edges), this sum- mation can be performed in polynomial time evaluat- The partition function of a graphical model, which ing the Pfaffian of a skew-symmetric matrix associated plays the role of normalization constant in a MRF or with the extended graph. Furthermore, the full loop probability of evidence (likelihood) in a BN is a fun- series can be expressed as a sum over Pfaffian terms, damental quantity which arises in many contexts such each one accounting for a large number of loops and of variables where σa is the vector of variables asso- σ solvable in polynomial time as well. ciated with node a, i.e. a := (σab1 ,σab2 ,... ) where b a¯. The joint probability distribution of such a This approach builds on classical results from 1960s i model∈ factorizes as: by Kasteleyn (1963); Fisher (1966) and other physi- −1 cists who showed that in a planar graphical model de- p (σ) = Z fa (σa), Z = fa (σa), (1) fined in terms of binary variables, computing Z can σ aY∈V X aY∈V be mapped to a weighted perfect matching problem where Z is the partition function. and calculated in polynomial time under the key re- striction that interactions only depend on agreement From a variational perspective, a fixed point of the BP or disagreement between the signs of their variables. algorithm represents a stationary point of the Bethe Such a model is known in statistical physics as the ”free energy” approximation under proper constraints Ising model without external field. Notice that exact (Yedidia et al., 2005): inference for a general binary graphical model on a ZBP = exp F BP , (2) planar graph (i.e. Ising model with external field) is − b (σ ) intractable (Barahona, 1982). F BP = b (σ ) ln a a a a f (σ ) a σ a a Recently, some methods for inference over graphical X Xa models, based on the works of Kasteleyn and Fisher, b (σ ) lnb (σ ), (3) − ab ab ab ab have been introduced. Globerson and Jaakkola (2007) b∈a¯ σab obtained upper bounds on Z for non-planar graphs X X σ with binary variables by decomposition of Z into a where ba( a) and bab(σab) are the beliefs (pseudo- marginals) associated to each node a and variable weighted sum over partition functions of spanning ∈ V tractable (zero field) planar models. Another example ab. For graphs without loops, Equation (2) coincides BP is the work of Schraudolph and Kamenetsky (2009) with the Gibbs ”free energy” and therefore Z coin- BP which provides a framework for exact inference on a cides with Z. If the graph contains loops, Z is just restricted class of planar graphs using the approach of an approximation critically dependent on how strong Kasteleyn and Fisher. the influence of the loops is. We introduce now some convenient definitions. Contrary to the two aforementioned approaches which rely on exact inference on a tractable planar model, the Definition 1 A generalized loop in a graph is any G loop calculus directly leads to a framework for approx- subgraph C such that each node in C has degree 2 or imate inference on general planar graphs. Truncating larger. the loop series according to Chertkov et al. (2008) al- ready gives the exact result in the zero external field We use the term ”loop” instead of ”generalized loop” case. In the general planar case, however, this trunca- for the rest of the manuscript. Z is explicitly repre- tion may result into an accurate approximation that sented in terms of the BP solution via the loop series can be incrementally corrected by considering subse- expansion: quent terms in the series. BP Z = Z z, z = 1 + rC , rC = µa;¯aC , · ! CX∈C aY∈C 2 Belief Propagation and loop Series (4) for Planar Graphs where is the set of all the loops within the graph. Each loopC term r is a product of terms µ asso- We consider the Forney graph representation, also C a,a¯C ciated with every node a of the loop.a ¯ denotes the called general vertex model (Forney, 2001; Loeliger, C set of neighbors of a within the loop C: 2004), of a probability distribution p(σ) defined over a vector σ of binary variables (vectors are denoted us- ba (σa) (σab mab) ing bold symbols). Forney graphs are associated with − σa b∈a¯C general graphical models which subsume other factor µa;¯aC = X Y , 1 m2 graphs, e.g. those correspondent to BNs and MRFs. − ab b∈a¯C q A binary Forney graph := ( , ) consists of a set of Y nodes where each nodeG a V representsE an interac- mab = σabbab (σab). (5) V ∈ V σab tion and each edge (a, b) represents a binary vari- X able ab which take values∈σ E := 1 . We denotea ¯ We consider planar graphs with all nodes of degree not ab {± } the set of neighbors of node a. Interactions fa (σa) are larger than 3, i.e. a¯C 3. We denote by triplet a arbitrary functions defined over typically small subsets node with degree 3| in the| ≤ graph. (a) a G g (b) Gext G Gext b f b c d e 1 i j h l a µa;{b,c} 1 k c (c) b 1 µ µ a a;{b,c} a;{b,d} 1 1 µa;{c,d} 1 2 3 (d) c d 4 5 Figure 2: Fisher’s rules. (Top) A node a of degree 2 in is split in 2 nodes in . (Bottom) A node a of G Gext degree 3 in is split in 3 nodes in ext. Right boxes include all matchingsG in relatedG with node a in . Gext G Figure 1: Example. (a) A Forney graph. (b) Cor- responding extended graph. (c) Loops (in bold) in- cluded in the 2-regular partition function Z∅. (d) otherwise. The following identity allows to obtain the Loops (in bold and red) not included in Z∅. Marked in Pfaffian up to a sign by computing the determinant: red, the triplets associated with each loop. Grouped in 2 gray squares, the loops considered in different subsets Pfaffian (A) = Det(A).
Load more

Recommended publications
  • Matchgates Revisited
    THEORY OF COMPUTING, Volume 10 (7), 2014, pp. 167–197 www.theoryofcomputing.org RESEARCH SURVEY Matchgates Revisited Jin-Yi Cai∗ Aaron Gorenstein Received May 17, 2013; Revised December 17, 2013; Published August 12, 2014 Abstract: We study a collection of concepts and theorems that laid the foundation of matchgate computation. This includes the signature theory of planar matchgates, and the parallel theory of characters of not necessarily planar matchgates. Our aim is to present a unified and, whenever possible, simplified account of this challenging theory. Our results include: (1) A direct proof that the Matchgate Identities (MGI) are necessary and sufficient conditions for matchgate signatures. This proof is self-contained and does not go through the character theory. (2) A proof that the MGI already imply the Parity Condition. (3) A simplified construction of a crossover gadget. This is used in the proof of sufficiency of the MGI for matchgate signatures. This is also used to give a proof of equivalence between the signature theory and the character theory which permits omittable nodes. (4) A direct construction of matchgates realizing all matchgate-realizable symmetric signatures. ACM Classification: F.1.3, F.2.2, G.2.1, G.2.2 AMS Classification: 03D15, 05C70, 68R10 Key words and phrases: complexity theory, matchgates, Pfaffian orientation 1 Introduction Leslie Valiant introduced matchgates in a seminal paper [24]. In that paper he presented a way to encode computation via the Pfaffian and Pfaffian Sum, and showed that a non-trivial, though restricted, fragment of quantum computation can be simulated in classical polynomial time. Underlying this magic is a way to encode certain quantum states by a classical computation of perfect matchings, and to simulate certain ∗Supported by NSF CCF-0914969 and NSF CCF-1217549.
    [Show full text]
  • Computations in Algebraic Geometry with Macaulay 2
    Computations in algebraic geometry with Macaulay 2 Editors: D. Eisenbud, D. Grayson, M. Stillman, and B. Sturmfels Preface Systems of polynomial equations arise throughout mathematics, science, and engineering. Algebraic geometry provides powerful theoretical techniques for studying the qualitative and quantitative features of their solution sets. Re- cently developed algorithms have made theoretical aspects of the subject accessible to a broad range of mathematicians and scientists. The algorith- mic approach to the subject has two principal aims: developing new tools for research within mathematics, and providing new tools for modeling and solv- ing problems that arise in the sciences and engineering. A healthy synergy emerges, as new theorems yield new algorithms and emerging applications lead to new theoretical questions. This book presents algorithmic tools for algebraic geometry and experi- mental applications of them. It also introduces a software system in which the tools have been implemented and with which the experiments can be carried out. Macaulay 2 is a computer algebra system devoted to supporting research in algebraic geometry, commutative algebra, and their applications. The reader of this book will encounter Macaulay 2 in the context of concrete applications and practical computations in algebraic geometry. The expositions of the algorithmic tools presented here are designed to serve as a useful guide for those wishing to bring such tools to bear on their own problems. A wide range of mathematical scientists should find these expositions valuable. This includes both the users of other programs similar to Macaulay 2 (for example, Singular and CoCoA) and those who are not interested in explicit machine computations at all.
    [Show full text]
  • On the Computation of Pfaffians?
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector DISCRETE APPLIED MATHEMATICS ELSEVIER Discrete Applied Mathematics 51 (1994) 269-275 On the computation of pfaffians? G. Galbiati” and F. Maffiolibs* “Universitir di Pavia, Italy bPolitecnico di Milano, Milano, Italy (Received 21 October 1991) Abstract We present an efficient algorithm for computing the pfaffian of a matrix whose elements belong to an integral domain. Relevant applications are exact value problems in matching and matroid theory. 1. Introduction This work deals with efficient ways of computing the pfaffian of a skew-symmetric matrix whose elements belong to an integral domain. Pfaffians play an important role in matching problems [9], and in their natural generalization into matroid parity problems [3,8]. Since for every skew-symmetric matrix of even order it is well known [7] that its determinant is equal to the square of the pfaffian of the matrix, in the solution of existence versions of the above mentioned matching and matroid parity problems, the computation of the pfaffian can be substituted by the computation of the determinant. However, in the solution of the corresponding exact value problems, computing pfaffians becomes essential [2,3]. The possibility of directly computing pfaffians was already pointed out in [l]. This paper presents an efficient algorithm for computing the pfaffian of a matrix whose elements belong to an integral domain. When applied to an integral matrix, similarly to Edmonds’ algorithm for computing the determinant [4], this algorithm works with elements of bounded magnitude, namely bounded above by the magni- tude of any minor of the given matrix.
    [Show full text]
  • An Exploration of the Permanent-Determinant Method
    An exploration of the permanent-determinant method Greg Kuperberg UC Davis [email protected] Abstract The permanent-determinant method and its generalization, the Hafnian- Pfaffian method, are methods to enumerate perfect matchings of plane graphs that were discovered by P. W. Kasteleyn. We present several new techniques and arguments related to the permanent-determinant with consequences in enu- merative combinatorics. Here are some of the results that follow from these techniques: 1. If a bipartite graph on the sphere with 4n vertices is invariant under the antipodal map, the number of matchings is the square of the number of matchings of the quotient graph. 2. The number of matchings of the edge graph of a graph with vertices of degree at most 3 is a power of 2. 3. The three Carlitz matrices whose determinants count a × b × c plane partitions all have the same cokernel. 4. Two symmetry classes of plane partitions can be enumerated with almost no calculation. Submitted: October 16, 1998; Accepted: November 9, 1998 [Also available as math.CO/9810091] The permanent-determinant method and its generalization, the Hafnian-Pfaffian method, is a method to enumerate perfect matchings of plane graphs that was dis- covered by P. W. Kasteleyn [18]. Given a bipartite plane graph Z, the method pro- duces a matrix whose determinant is the number of perfect matchings of Z.Given a non-bipartite plane graph Z, it produces a Pfaffian with the same property. The method can be used to enumerate symmetry classes of plane partitions [21, 22] and domino tilings of an Aztec diamond [45] and is related to some recent factorizations of the number of matchings of plane graphs with symmetry [5, 15].
    [Show full text]
  • Holographic Algorithms
    Holographic Algorithms Jin-Yi Cai ∗ Computer Sciences Department University of Wisconsin Madison, WI 53706. USA. Email: [email protected] Abstract Leslie Valiant recently proposed a theory of holographic algorithms. These novel algorithms achieve exponential speed-ups for certain computational problems compared to naive algorithms for the same problems. The methodology uses Pfaffians and (planar) perfect matchings as basic computational primitives, and attempts to create exponential cancellations in computation. In this article we survey this new theory of matchgate computations and holographic algorithms. Key words: Theoretical Computer Science, Computational Complexity Theory, Perfect Match- ings, Pfaffians, Matchgates, Matchcircuits, Matchgrids, Signatures, Holographic Algorithms. 1 Some Historical Background There have always been two major strands of mathematical thought since antiquity and across civilizations: Structural Theory and Computation, as exemplified by Euclid’s Elements and Dio- phantus’ Arithmetica. Structural Theory prizes the formulation and proof of structural theorems, while Computation seeks efficient algorithmic methods to solve problems. Of course, these strands of mathematical thought are not in opposition to each other, but rather they are highly intertwined and mutually complementary. For example, from Euclid’s Elements we learn the Euclidean algo- rithm to find the greatest common divisor of two positive integers. This algorithm can serve as the first logical step in the structural derivation of elementary number theory. At the same time, the correctness and efficiency of this and similar algorithms demand proofs in a purely structural sense, and use quite a bit more structural results from number theory [4]. As another example, the computational difficulty of recognizing primes and the related (but separate) problem of in- teger factorization already fascinated Gauss, and are closely tied to the structural theory of the distribution of primes [21, 1, 2].
    [Show full text]
  • Pfaffian Schur Processes and Last Passage Percolation in a Half-Quadrant
    PFAFFIAN SCHUR PROCESSES AND LAST PASSAGE PERCOLATION IN A HALF-QUADRANT JINHO BAIK, GUILLAUME BARRAQUAND, IVAN CORWIN, AND TOUFIC SUIDAN Abstract. We study last passage percolation in a half-quadrant, which we analyze within the framework of Pfaffian Schur processes. For the model with exponential weights, we prove that the fluctuations of the last passage time to a point on the diagonal are either GSE Tracy-Widom distributed, GOE Tracy-Widom distributed, or Gaussian, depending on the size of weights along the diagonal. Away from the diagonal, the fluctuations of passage times follow the GUE Tracy- Widom distribution. We also obtain a two-dimensional crossover between the GUE, GOE and GSE distribution by studying the multipoint distribution of last passage times close to the diagonal when the size of the diagonal weights is simultaneously scaled close to the critical point. We expect that this crossover arises universally in KPZ growth models in half-space. Along the way, we introduce a method to deal with diverging correlation kernels of point processes where points collide in the scaling limit. Contents 1. Introduction 1 2. Fredholm Pfaffian formulas8 3. Pfaffian Schur process 13 4. Fredholm Pfaffian formulas for k-point distributions 23 5. Convergence to the GSE Tracy-Widom distribution 33 6. Asymptotic analysis in other regimes: GUE, GOE, crossovers 46 References 56 1. Introduction 2 In this paper, we study last passage percolation in a half-quadrant of Z . We extend all known results for the case of geometric weights (see discussion of previous results below) to the case of exponential weights.
    [Show full text]
  • The Tutte Polynomial and Related Polynomials Lecture Notes 2010, 2012, 2014
    The Tutte polynomial and related polynomials Lecture notes 2010, 2012, 2014 Andrew Goodall The following notes derive from three related series of lectures given for the Selected Chapters in Combinatorics course (Vybran´ekapitoly z kombinatoriky I) at the Computer Science Institute (IUUK)´ and the Department of Applied Mathematics (KAM) of the Faculty of Mathematics and Physics (MFF) at Charles University, Prague: \Many facets of the Tutte polynomial" (2010), \Graph invariants, homomorphisms and the Tutte polynomial" (2012), and \Duality in combinatorics: the examples of Tutte, Erd}os,and Ramsey" (2014).1 I have tried to make the notation as uniform as possible throughout and to avoid repetitions arising from overlaps between the three courses. However, I have left some background to cycles and cuts in Section 6 as it stands, rather than asking the reader to find the relevant material in Section 2. Contents 1 The chromatic polynomial 1 1.1 Graph-theoretic preliminaries . 1 1.2 The chromatic polynomial and proper colourings . 2 1.3 Deletion and contraction . 6 1.4 Subgraph expansions . 12 1.5 Some other deletion{contraction invariants. 15 2 Flows and tensions 17 2.1 Orientations . 17 2.2 Circuits and cocircuits . 18 2.3 The incidence matrix of an oriented graph . 21 2.4 A-flows and A-tensions . 23 2.5 Tensions and colourings . 26 2.6 Duality of bases for A-tensions and A-flows . 28 2.7 Examples of nowhere-zero flows . 29 2.8 The flow polynomial . 33 1The courses in 2012 and 2014 were complemented by lectures given by Prof.
    [Show full text]
  • Pfaffians and Perfect Matchings? 1. Matchings Given a Graph GV,E With
    What are… Pfaffians and Perfect Matchings? 1. Matchings Given a graph ! !, ! with vertex set ! and edge set !, a matching is a subset ! ⊆ ! such that no two edges in ! share a common vertex. A perfect matching is a matching in which every vertex of ! is met by an edge. We wish to develop a determinantial formula for the generating function of perfect matchings in a graph. 2. The Pfaffian and Skew Symmetric Matrices In 1882 Thomas Muir proved that the determinant of a skew matrix is the square of a polynomial, and noted that this polynomial was the Pfaffian Caley mentioned in 1852 in his essay “On the theory of permutants.” A formal definition for the determinant of an !×! matrix is given by ! !"# ! = !"# ! !!,! ! !∈!! !!! For a 2!×2! skew matrix (each entry is the opposite of its conjugate) a formal definition of the Pfaffian may be adapted from that of the determinant and is given by ! 1 !" ! = !"# ! ! 2!!! ! !!!! ,! !! !∈!!! !!! However, most of these terms are redundant for our purposes. For example, if we swap 2! − 1 with 2! − 1 and 2! with 2! we effectively permute the factors !!,!. Reordering the factors !!,! does not change the sign of ! because it is an even number of swaps from one permutation ! to another. There are !! such permutations of the factors !!,!. Additionally, the order of the indices of each !!,! may be reversed. This is a single swap in the permutation, so the sign of ! switches. However, since the matrix is skew ! !!,! = −!!,! and thus the sign of the term as a whole remains constant. There are 2 ways to order all the indices together.
    [Show full text]
  • 2D Tiling: Domino, Tetromino and Percolation
    2D Tiling: Domino, Tetromino and Percolation D. V. Chudnovsky, G.V. Chudnovsky IMAS NYU Polytechnic School of Engineering 6MetroTechCenter Brooklyn, NY 11201 March 5, 2015 March 26, 2015 → General Definitions Let Γbe a finite connected (undirected) graph, and let V (Γ) and E(Γ) be, respectively, the vertices and edges of Γ. A perfect matching on Γis a set of edges of Γsuch that every vertex is adjacent to exactly one of the edges in the matching. If there is no perfect matching on Γone uses a maximum matching as a largest in cardinality. The existence of a perfect matching, as well as finding a maximum matching in a graph Γcan be determined in a polynomial time. However, the problem of finding the number of perfect matching is #P for general graphs Γ. One of the most common cases is the case of graphs on a square lattice. One gets a large chessboard and ”domino covering problem”. In the case of planar graphs the counting of dimer coverings (Γ) can be D effectively computed using techniques developed by physicists in early 1960s. If A =(aij ) is a skew-symmetric matrix, then its determinant, det(A)isa square of a polynomial in aij which is called a Pfaffian, Pf(A), of A. FKT Formula and Kasteleyn Ordering of Planar Graphs Fisher, Kasteleyn, and Temperley (FKT) in 1961, established an efficient algorithm of computing the number of dimer coverings for any planar graph Γ using Pfaffians and special orientations of edges of Γ. To define the corresponding matrices, let us look at an orientation K of edges K K Γ.
    [Show full text]
  • Arxiv:1606.00525V4 [Math.PR] 9 Jan 2020 Es Rdompaa,Paetransition
    PFAFFIAN SCHUR PROCESSES AND LAST PASSAGE PERCOLATION IN A HALF-QUADRANT JINHO BAIK, GUILLAUME BARRAQUAND, IVAN CORWIN, AND TOUFIC SUIDAN Abstract. We study last passage percolation in a half-quadrant, which we analyze within the framework of Pfaffian Schur processes. For the model with exponential weights, we prove that the fluctuations of the last passage time to a point on the diagonal are either GSE Tracy-Widom distributed, GOE Tracy-Widom distributed, or Gaussian, depending on the size of weights along the diagonal. Away from the diagonal, the fluctuations of passage times follow the GUE Tracy- Widom distribution. We also obtain a two-dimensional crossover between the GUE, GOE and GSE distribution by studying the multipoint distribution of last passage times close to the diagonal when the size of the diagonal weights is simultaneously scaled close to the critical point. We expect that this crossover arises universally in KPZ growth models in half-space. Along the way, we introduce a method to deal with diverging correlation kernels of point processes where points collide in the scaling limit. Contents 1. Introduction 1 2. Fredholm Pfaffian formulas 8 3. Pfaffian Schur process 13 4. Fredholm Pfaffian formulas for k-point distributions 23 5. Convergence to the GSE Tracy-Widom distribution 33 6. Asymptotic analysis in other regimes: GUE, GOE, crossovers 46 References 56 1. Introduction In this paper, we study last passage percolation in a half-quadrant of Z2. We extend all known arXiv:1606.00525v4 [math.PR] 9 Jan 2020 results for the case of geometric weights (see discussion of previous results below) to the case of exponential weights.
    [Show full text]
  • The Pfaffian Transformation
    The Pfaffian Transformation Tracale Austin, Hans Bantilan, Isao Jonas, and Paul Kory March 9, 2007 Abstract We define a function on sequences, which we call the Pfaffian transformation, based on the concept of the Pfaffian of a skew-symmetric matrix. We discuss prop- erties of the Pfaffian transformation, culminating in a theorem proving that the Pfaffian transformation maps sequences satisfying linear recurrences to sequences satisfying linear recurrences. 1 Introduction to the Pfaffian Transformation Given a sequence (a0, a1, a2, ...), we define a sequence of skew-symmetric matrices (A0, A1, A2, ...) 0 a0 a1 a2 ··· a2n −a 0 a a ··· a 0 0 1 2n−1 −a1 −a0 0 a0 ··· a2n−2 where An is −a −a −a 0 ··· a . 2 1 0 2n−3 . . .. −a2n −a2n−1 −a2n−2 −a2n−3 ··· 0 We then define the Pfaffian transformation of (a0, a1, a2, ...) to be the sequence of Pfaffians (P f(A0), P f(A1), P f(A2), ...), where the Pfaffian of a skew-symmetric matrix is the positive or negative square root of its determinant. (For a precise definition of the Pfaffian, see section 2.) The Pfaffian transformation is thus a function from sequences to sequences. We begin by observing the effects of this transformation on some simple, well-known sequences: (1, 2, 3, 4, 5, ...) → (1, 0, 0, 0, 0, ...) (2, 2, 2, 2, 2, ...) → (2, 4, 8, 16, 32, ...) (1, 2, 4, 8, 16, ...) → (1, 1, 1, 1, 1, ...) (1, 1, 2, 3, 5, ...) → (1, 2, 4, 8, 16, ...) Observe that each of these input sequences satisfies a linear recurrence relation and is mapped to a sequence that also satisfies a linear recurrence.
    [Show full text]
  • Pfaffian Schur Processes and Last Passage Percolation in a Half-Quadrant1
    The Annals of Probability 2018, Vol. 46, No. 6, 3015–3089 https://doi.org/10.1214/17-AOP1226 © Institute of Mathematical Statistics, 2018 PFAFFIAN SCHUR PROCESSES AND LAST PASSAGE PERCOLATION IN A HALF-QUADRANT1 ∗ BY JINHO BAIK ,2,GUILLAUME BARRAQUAND†,3,IVA N CORWIN†,4 AND TOUFIC SUIDAN University of Michigan∗ and Columbia University† We study last passage percolation in a half-quadrant, which we analyze within the framework of Pfaffian Schur processes. For the model with expo- nential weights, we prove that the fluctuations of the last passage time to a point on the diagonal are either GSE Tracy–Widom distributed, GOE Tracy– Widom distributed or Gaussian, depending on the size of weights along the diagonal. Away from the diagonal, the fluctuations of passage times fol- low the GUE Tracy–Widom distribution. We also obtain a two-dimensional crossover between the GUE, GOE and GSE distribution by studying the mul- tipoint distribution of last passage times close to the diagonal when the size of the diagonal weights is simultaneously scaled close to the critical point. We expect that this crossover arises universally in KPZ growth models in half-space. Along the way, we introduce a method to deal with diverging cor- relation kernels of point processes where points collide in the scaling limit. CONTENTS 1. Introduction ............................................3016 1.1. Previous results in (half-space) LPP ............................3017 1.2. Previous methods ......................................3018 1.3. Our methods and new results ................................3018 1.4.OthermodelsrelatedtoKPZgrowthinahalf-space...................3020 1.5.Mainresults.........................................3020 Outline of the paper ........................................3024 2. Fredholm Pfaffian formulas ...................................3024 2.1.GUETracy–Widomdistribution..............................3024 2.2.
    [Show full text]