DOCSLIB.ORG

Statistical Algorithms and a Lower Bound for Detecting Planted Cliques

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Statistical Algorithms and a Lower Bound for Detecting Planted Cliques Statistical Algorithms and a Lower Bound for Detecting Planted Cliques ∗ y z y Vitaly Feldman Elena Grigorescu Lev Reyzin Almaden Research Center Dept. of Computer Science Dept. of Mathematics IBM Purdue University University of Illinois at Chicago San Jose, CA 95120 West Lafayette, IN 47907 Chicago, IL 60607 [email protected] [email protected] [email protected] y y Santosh S. Vempala Ying Xiao School of Computer Science School of Computer Science Georgia Inst. of Technology Georgia Inst. of Technology Atlanta, GA 30332 Atlanta, GA 30332 [email protected] [email protected] ABSTRACT Categories and Subject Descriptors We introduce a framework for proving lower bounds on com- F.2 [Analysis of Algorithms and Problem Complex- putational problems over distributions, based on a class of ity]: General algorithms called statistical algorithms. For such algorithms, access to the input distribution is limited to obtaining an es- Keywords timate of the expectation of any given function on a sample drawn randomly from the input distribution, rather than Statistical algorithms; Lower bounds; Planted clique; Sta- directly accessing samples. Most natural algorithms of in- tistical query terest in theory and in practice, e.g., moments-based meth- ods, local search, standard iterative methods for convex op- 1. INTRODUCTION timization, MCMC and simulated annealing, are statistical We study the complexity of problems where the input algorithms or have statistical counterparts. Our framework consists of independent samples from an unknown distri- is inspired by and generalize the statistical query model in bution. Such problems are at the heart of machine learning learning theory [34]. and statistics (and their numerous applications) and also oc- Our main application is a nearly optimal lower bound cur in many other contexts such as compressed sensing and on the complexity of any statistical algorithm for detect- cryptography. Many methods exist to estimate the sample ing planted bipartite clique distributions (or planted dense complexity of such problems (e.g. VC dimension [47]). Prov- subgraph distributions) when the planted clique has size ing lower bounds on the computational complexity of these 1=2−δ O(n ) for any constant δ > 0. Variants of these prob- problems has been much more challenging. The traditional lems have been assumed to be hard to prove hardness for approach to this is via reductions and finding distributions other problems and for cryptographic applications. Our that can generate instances of some problem conjectured to lower bounds provide concrete evidence of hardness, thus be intractable (e.g., assuming NP 6= RP). supporting these assumptions. Here we present a different approach, namely showing that a broad class of algorithms, which we refer to as statistical al- ∗This material is based upon work supported by the Na- gorithms, have high complexity, unconditionally. Our defini- tional Science Foundation under Grant #1019343 to the tion encompasses most algorithmic approaches used in prac- Computing Research Association for the CIFellows Project. tice and in theory on a wide variety of problems, including yResearch supported in part by NSF awards CCF-0915903 Expectation Maximization (EM) [16], local search, MCMC and CCF-0910584. optimization [45, 25], simulated annealing [36, 48], first and zResearch supported in part by a Simons Postdoctoral Fel- second order methods for linear/convex optimization, [17, lowship. 6], k-means, Principal Component Analysis (PCA), Inde- pendent Component Analysis (ICA), Na¨ıve Bayes, Neural Networks and many others (see [13] and [9] for proofs and many other examples). In fact, we are aware of only one al- Permission to make digital or hard copies of all or part of this work for gorithm that does not have a statistical counterpart: Gaus- personal or classroom use is granted without fee provided that copies are sian elimination for solving linear equations over a field (e.g., not made or distributed for profit or commercial advantage and that copies mod 2). bear this notice and the full citation on the first page. To copy otherwise, to Informally, statistical algorithms can access the input dis- republish, to post on servers or to redistribute to lists, requires prior specific tribution only by asking for the value of any bounded real- permission and/or a fee. STOC’13, June 1-4, 2013, Palo Alto, California, USA. valued function on a random sample from the distribution, Copyright 2013 ACM 978-1-4503-2029-0/13/06 ...$15.00. or the average value of the function over a specified number of independent random samples. As an example, suppose lem of finding the largest clique in a random graph [33]. A we are trying to solve minu2U Ex∼D[f(x; u)] by gradient de- random graph Gn;1=2 contains a clique of size 2 log n with scent. Then the gradient of the objective function is (by high probability, and a simple greedy algorithm can find one interchanging the derivative with the integral) of size log n. Finding cliques of size (2 − ) log n is a hard problem for any > 0. Planting a larger clique should make ru E[f(x; u)] = E[ru(f(x; u))] x x it easier to find one. The problem of finding the smallest k for which the planted clique can be detected in polynomialp and can be estimated from samples; thus the algorithm can time has attracted significant attention. For k ≥ c n log n, proceed without ever examining samples directly. A more simply picking verticesp of large degrees suffices [37]. Cliques involved example is Linear Programming. One version is the of size k = Ω( n) can be found using spectral methods feasibility problem: find a nonzero w s.t. a · w ≥ 0 for all a [2, 38, 14], via SDPs [19], nuclear norm minimization [3] or in some set A. We can formulate this as combinatorial methods [21, 15]. While there is no known polynomial-time algorithm that max E[sign(a · w)] p w a can detect cliques of size below the threshold of Ω( n), there is a quasipolynomial algorithm for any k ≥ 2 log n: enumer- and the distribution over a could be uniform over the set A ate subsets of size 2 log n; for each subset that forms a clique, if it is finite. This can be solved by a statistical algorithm take all common neighbors of the subset; one of these will be [10, 17]. This is also the case for semidefinite programs and the planted clique. This is also the fastest known algorithm for conic optimization [6]. The key motivation for our def- for any k = O(n1=2−δ), where δ > 0. inition of statistical algorithms is the empirical observation Some evidence of the hardness of the problem was shown that almost all algorithms that work on explicit instances by Jerrum [30] who proved that a specific approach using are already statistical in our sense or have natural statisti- p a Markov chain cannot be efficient for k = o( n). More cal counterparts. Thus, proving lower bounds for statistical evidence of hardness is given in [20], where it is shown that algorithms strongly indicates the need for new approaches Lov´asz-Schrijver SDP relaxations, which include the SDP even for explicit instances. We present the formal oracle- used in [19], cannot be used to efficiently find cliques of based definitions of statistical algorithms in Section 2. p size k = o( n). The problem has been used to generate The inspiration for our model is the statistical query (SQ) cryptographic primitives [31], and as a hardness assumption model in learning theory [34] defined as a restriction of [1, 28, 40]. Valiant's PAC learning model [46]. The primary goal of We focus on the bipartite planted clique problem, where a the restriction was to simplify the design of noise-tolerant bipartite k ×k clique is planted in a random bipartite graph. learning algorithms. As was shown by Kearns and others A densest-subgraph version of the bipartite planted clique in subsequent works, almost all classes of functions that can problem has been used as a hard problem for cryptographic be learned efficiently can also be efficiently learned in the applications [4]. All known bounds and algorithms for the restricted SQ model. A notable and so far the only excep- k-clique problem can be easily adapted to the bipartite case. tion is the algorithm for learning parities, based on Gaussian Therefore it is natural to suspect that new upper bounds on elimination. As was already shown by Kearns [34], parities the planted k-clique problem would also yield new upper require exponential time to learn in the SQ model. Further, bounds for the bipartite case. Blum et al. [11] proved that the number of SQs required The starting point of our investigation for this problem for weak learning (that is, for obtaining a non-negligible ad- is the property of the bipartite planted k-clique problem vantage over the random guessing) of a class of functions C is that it has an equivalent formulation as a problem over over a fixed distribution D is characterized by a combina- distributions defined as follows. torial parameter of C and D, referred to as SQ-DIM(C; D), the statistical query dimension. Problem 1. Fix an integer k, 1 ≤ k ≤ n, and a subset Our notion of statistical algorithms generalizes SQ learn- of k indices S ⊆ f1; 2; : : : ; ng. The input distribution DS ing algorithms to any computational problem over distribu- on vectors x 2 f0; 1gn is defined as follows: with probability tions. For any problem over distributions we define a pa- 1 − (k=n), x is uniform over f0; 1gn; and with probability rameter of the problem that lower bounds the complexity of k=n the k coordinates of x in S are set to 1, and the rest solving the problem by any statistical algorithm in the same are uniform over their support.
Load more

Recommended publications
  • The Strongish Planted Clique Hypothesis and Its Consequences
    The Strongish Planted Clique Hypothesis and Its Consequences Pasin Manurangsi Google Research, Mountain View, CA, USA [email protected] Aviad Rubinstein Stanford University, CA, USA [email protected] Tselil Schramm Stanford University, CA, USA [email protected] Abstract We formulate a new hardness assumption, the Strongish Planted Clique Hypothesis (SPCH), which postulates that any algorithm for planted clique must run in time nΩ(log n) (so that the state-of-the-art running time of nO(log n) is optimal up to a constant in the exponent). We provide two sets of applications of the new hypothesis. First, we show that SPCH implies (nearly) tight inapproximability results for the following well-studied problems in terms of the parameter k: Densest k-Subgraph, Smallest k-Edge Subgraph, Densest k-Subhypergraph, Steiner k-Forest, and Directed Steiner Network with k terminal pairs. For example, we show, under SPCH, that no polynomial time algorithm achieves o(k)-approximation for Densest k-Subgraph. This inapproximability ratio improves upon the previous best ko(1) factor from (Chalermsook et al., FOCS 2017). Furthermore, our lower bounds hold even against fixed-parameter tractable algorithms with parameter k. Our second application focuses on the complexity of graph pattern detection. For both induced and non-induced graph pattern detection, we prove hardness results under SPCH, improving the running time lower bounds obtained by (Dalirrooyfard et al., STOC 2019) under the Exponential Time Hypothesis. 2012 ACM Subject Classification Theory of computation → Problems, reductions and completeness; Theory of computation → Fixed parameter tractability Keywords and phrases Planted Clique, Densest k-Subgraph, Hardness of Approximation Digital Object Identifier 10.4230/LIPIcs.ITCS.2021.10 Related Version A full version of the paper is available at https://arxiv.org/abs/2011.05555.
    [Show full text]
  • Computational Lower Bounds for Community Detection on Random Graphs
    JMLR: Workshop and Conference Proceedings vol 40:1–30, 2015 Computational Lower Bounds for Community Detection on Random Graphs Bruce Hajek [email protected] Department of ECE, University of Illinois at Urbana-Champaign, Urbana, IL Yihong Wu [email protected] Department of ECE, University of Illinois at Urbana-Champaign, Urbana, IL Jiaming Xu [email protected] Department of Statistics, The Wharton School, University of Pennsylvania, Philadelphia, PA, Abstract This paper studies the problem of detecting the presence of a small dense community planted in a large Erdos-R˝ enyi´ random graph G(N; q), where the edge probability within the community exceeds q by a constant factor. Assuming the hardness of the planted clique detection problem, we show that the computational complexity of detecting the community exhibits the following phase transition phenomenon: As the graph size N grows and the graph becomes sparser according to −α 2 q = N , there exists a critical value of α = 3 , below which there exists a computationally intensive procedure that can detect far smaller communities than any computationally efficient procedure, and above which a linear-time procedure is statistically optimal. The results also lead to the average-case hardness results for recovering the dense community and approximating the densest K-subgraph. 1. Introduction Networks often exhibit community structure with many edges joining the vertices of the same com- munity and relatively few edges joining vertices of different communities. Detecting communities in networks has received a large amount of attention and has found numerous applications in social and biological sciences, etc (see, e.g., the exposition Fortunato(2010) and the references therein).
    [Show full text]
  • Spectral Algorithms (Draft)
    Spectral Algorithms (draft) Ravindran Kannan and Santosh Vempala March 3, 2013 ii Summary. Spectral methods refer to the use of eigenvalues, eigenvectors, sin- gular values and singular vectors. They are widely used in Engineering, Ap- plied Mathematics and Statistics. More recently, spectral methods have found numerous applications in Computer Science to \discrete" as well \continuous" problems. This book describes modern applications of spectral methods, and novel algorithms for estimating spectral parameters. In the first part of the book, we present applications of spectral methods to problems from a variety of topics including combinatorial optimization, learning and clustering. The second part of the book is motivated by efficiency considerations. A fea- ture of many modern applications is the massive amount of input data. While sophisticated algorithms for matrix computations have been developed over a century, a more recent development is algorithms based on \sampling on the fly” from massive matrices. Good estimates of singular values and low rank ap- proximations of the whole matrix can be provably derived from a sample. Our main emphasis in the second part of the book is to present these sampling meth- ods with rigorous error bounds. We also present recent extensions of spectral methods from matrices to tensors and their applications to some combinatorial optimization problems. Contents I Applications 1 1 The Best-Fit Subspace 3 1.1 Singular Value Decomposition . .3 1.2 Algorithms for computing the SVD . .7 1.3 The k-means clustering problem . .8 1.4 Discussion . 11 2 Clustering Discrete Random Models 13 2.1 Planted cliques in random graphs .
    [Show full text]
  • Public-Key Cryptography in the Fine-Grained Setting
    Public-Key Cryptography in the Fine-Grained Setting B B Rio LaVigne( ), Andrea Lincoln( ), and Virginia Vassilevska Williams MIT CSAIL and EECS, Cambridge, USA {rio,andreali,virgi}@mit.edu Abstract. Cryptography is largely based on unproven assumptions, which, while believable, might fail. Notably if P = NP, or if we live in Pessiland, then all current cryptographic assumptions will be broken. A compelling question is if any interesting cryptography might exist in Pessiland. A natural approach to tackle this question is to base cryptography on an assumption from fine-grained complexity. Ball, Rosen, Sabin, and Vasudevan [BRSV’17] attempted this, starting from popular hardness assumptions, such as the Orthogonal Vectors (OV) Conjecture. They obtained problems that are hard on average, assuming that OV and other problems are hard in the worst case. They obtained proofs of work, and hoped to use their average-case hard problems to build a fine-grained one-way function. Unfortunately, they proved that constructing one using their approach would violate a popular hardness hypothesis. This moti- vates the search for other fine-grained average-case hard problems. The main goal of this paper is to identify sufficient properties for a fine-grained average-case assumption that imply cryptographic prim- itives such as fine-grained public key cryptography (PKC). Our main contribution is a novel construction of a cryptographic key exchange, together with the definition of a small number of relatively weak struc- tural properties, such that if a computational problem satisfies them, our key exchange has provable fine-grained security guarantees, based on the hardness of this problem.
    [Show full text]
  • A New Approach to the Planted Clique Problem
    A new approach to the planted clique problem Alan Frieze∗ Ravi Kannan† Department of Mathematical Sciences, Department of Computer Science, Carnegie-Mellon University. Yale University. [email protected]. [email protected]. 20, October, 2003 1 Introduction It is well known that finding the largest clique in a graph is NP-hard, [11]. Indeed, Hastad [8] has shown that it is NP-hard to approximate the size of the largest clique in an n vertex graph to within 1 ǫ a factor n − for any ǫ > 0. Not surprisingly, this has directed some researchers attention to finding the largest clique in a random graph. Let Gn,1/2 be the random graph with vertex set [n] in which each possible edge is included/excluded independently with probability 1/2. It is known that whp the size of the largest clique is (2+ o(1)) log2 n, but no known polymomial time algorithm has been proven to find a clique of size more than (1+ o(1)) log2 n. Karp [12] has even suggested that finding a clique of size (1+ ǫ)log2 n is computationally difficult for any constant ǫ > 0. Significant attention has also been directed to the problem of finding a hidden clique, but with only limited success. Thus let G be the union of Gn,1/2 and an unknown clique on vertex set P , where p = P is given. The problem is to recover P . If p c(n log n)1/2 then, as observed by Kucera [13], | | ≥ with high probability, it is easy to recover P as the p vertices of largest degree.
    [Show full text]
  • Lipics-CCC-2021-35.Pdf (1
    Hardness of KT Characterizes Parallel Cryptography Hanlin Ren # Ñ Institute for Interdisciplinary Information Sciences, Tsinghua University, Beijing, China Rahul Santhanam # University of Oxford, UK Abstract A recent breakthrough of Liu and Pass (FOCS’20) shows that one-way functions exist if and only if the (polynomial-)time-bounded Kolmogorov complexity, Kt, is bounded-error hard on average to compute. In this paper, we strengthen this result and extend it to other complexity measures: We show, perhaps surprisingly, that the KT complexity is bounded-error average-case hard if and only if there exist one-way functions in constant parallel time (i.e. NC0). This result crucially relies on the idea of randomized encodings. Previously, a seminal work of Applebaum, Ishai, and Kushilevitz (FOCS’04; SICOMP’06) used the same idea to show that NC0-computable one-way functions exist if and only if logspace-computable one-way functions exist. Inspired by the above result, we present randomized average-case reductions among the NC1- versions and logspace-versions of Kt complexity, and the KT complexity. Our reductions preserve both bounded-error average-case hardness and zero-error average-case hardness. To the best of our knowledge, this is the first reduction between the KT complexity and a variant of Kt complexity. We prove tight connections between the hardness of Kt complexity and the hardness of (the hardest) one-way functions. In analogy with the Exponential-Time Hypothesis and its variants, we define and motivate the Perebor Hypotheses for complexity measures such as Kt and KT. We show that a Strong Perebor Hypothesis for Kt implies the existence of (weak) one-way functions of near-optimal hardness 2n−o(n).
    [Show full text]
  • Spectral Barriers in Certification Problems
    Spectral Barriers in Certification Problems by Dmitriy Kunisky A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Mathematics New York University May, 2021 Professor Afonso S. Bandeira Professor Gérard Ben Arous © Dmitriy Kunisky all rights reserved, 2021 Dedication To my grandparents. iii Acknowledgements First and foremost, I am immensely grateful to Afonso Bandeira for his generosity with his time, his broad mathematical insight, his resourceful creativity, and of course his famous reservoir of open problems. Afonso’s wonderful knack for keeping up the momentum— for always finding some new angle or somebody new to talk to when we were stuck, for mentioning “you know, just yesterday I heard a very nice problem...” right when I could use something else to work on—made five years fly by. I have Afonso to thank for my sense of taste in research, for learning to collaborate, for remembering to enjoy myself, and above all for coming to be ever on the lookout for something new to think about. I am also indebted to Gérard Ben Arous, who, though I wound up working on topics far from his main interests, was always willing to hear out what I’d been doing, to encourage and give suggestions, and to share his truly encyclopedic knowledge of mathematics and its culture and history. On many occasions I’ve learned more tracking down what Gérard meant with the smallest aside in the classroom or in our meetings than I could manage to get out of a week on my own in the library.
    [Show full text]
  • Parameterized AC0 – Some Upper and Lower Bounds
    Parameterized AC0 { Some upper and lower bounds Yijia Chen Fudan University September 19th, 2017 @ Shonan AC0 0 A family of Boolean circuits Cn are AC -circuits if for every n 2 n2N N n (i)C n computes a Boolean function from f0; 1g to f0; 1g; (ii) the depth of Cn is bounded by a fixed constant; (iii) the size of Cn is polynomially bounded in n. Remark 1. Without (ii), we get a family of polynomial-size circuits Cn , which n2N decides a language in P/ poly. 2. If Cn is computable by a TM in time O(log n), then Cn is n2N dlogtime-uniform, which corresponds to FO(<; +; ×) [Barrington, Immerman, and Straubing, 1990]. The k-clique problem Definition Let G be a graph and k 2 N. Then a subset C ⊆ V (G) is a k-clique if (i) for every two vertices u; v 2 V (G) either u = v or fu; vg 2 E(G), (ii) and jCj = k. 296 CHAPTER 7 / TIME COMPLEXITY FIGURE 7.23 A graph with a 5-cliqueA graph with a 5-clique. The clique problem is to determine whether a graph contains a clique of a specified size. Let CLIQUE = G, k G is an undirected graph with a k-clique . {h i| } THEOREM 7.24 CLIQUE is in NP. PROOF IDEA The clique is the certificate. PROOF The following is a verifier V for CLIQUE. V = “On input G, k , c : hh i i 1. Test whether c is a subgraph with k nodes in G. 2.
    [Show full text]
  • Finding Planted Cliques in Sublinear Time
    Finding Planted Cliques in Sublinear Time Jay Mardia∗ Hilal Asi∗ Kabir Aladin Chandrasekher∗ Abstract We study the planted clique problem in which a clique of size k is planted in an Erd}os-R´enyi graph G(n; 1 ) and one is interested in recovering this planted clique. It is well known that for p 2 k = Ω( n), polynomial time algorithms can find the planted clique. In fact, the fastest known algorithms in this regime run in time linear O(n2) (or nearly linear) in the size of the input [FR10, DGGP14, DM15a]. In this work, we initiate thep development of sublinear time algorithms that find the planted clique in the regime k = !( n log log n). Our algorithms can reliably recover the clique n 3 3 p 2 k2 in time Oe n + ( ) = Oe n 2 when k = Ω( n log n), and in time Oe n = exp for p k p 24n !( n log log n) = k = o( n log n). An Ω(n) running time lower bound for the planted clique recovery problem follows easily from the results of [RS19] and therefore our recovery algorithms 2 are optimal whenever k = Ω(n 3 ). As the lower bound of [RS19] builds on purely information theoretic arguments, it cannot n2 p provide a detection lower bound stronger than Ω(e k2 ). Since our algorithms for k = Ω( n log n) n3 run in time Oe k3 + n , we show stronger lower bounds based on computational hardness assumptions. Using a slightly different formalization of the planted clique problem, in which every vertex is included in the clique independently with probability k=n, we show that the Planted Clique Conjecture implies the following.
    [Show full text]
  • How to Play Unique Games Against a Semi-Random Adversary
    How to Play Unique Games against a Semi-Random Adversary Study of Semi-Random Models of Unique Games Alexandra Kolla Konstantin Makarychev Yury Makarychev Microsoft Research IBM Research TTIC Abstract In this paper, we study the average case complexity of the Unique Games problem. We propose a natural semi-random model, in which a unique game instance is generated in several steps. First an adversary selects a completely satisfiable instance of Unique Games, then she chooses an "–fraction of all edges, and finally replaces (“corrupts”) the constraints corresponding to these edges with new constraints. If all steps are adversarial, the adversary can obtain any (1 − ") satisfiable instance, so then the problem is as hard as in the worst case. In our semi-random model, one of the steps is random, and all other steps are adversarial. We show that known algorithms for unique games (in particular, all algorithms that use the standard SDP relaxation) fail to solve semi-random instances of Unique Games. We present an algorithm that with high probability finds a solution satisfying a (1 − δ) fraction of all constraints in semi-random instances (we require that the average degree of the graph is Ω(log~ k)). To this end, we consider a new non-standard SDP program for Unique Games, which is not a relaxation for the problem, and show how to analyze it. We present a new rounding scheme that simultaneously uses SDP and LP solutions, which we believe is of independent interest. Our result holds only for " less than some absolute constant. We prove that if " ≥ 1=2, then the problem is hard in one of the models, that is, no polynomial–time algorithm can distinguish between the following two cases: (i) the instance is a (1 − ") satisfiable semi–random instance and (ii) the instance is at most δ satisfiable (for every δ > 0); the result assumes the 2–to–2 conjecture.
    [Show full text]
  • Inapproximability of Maximum Biclique Problems, Minimum $ K $-Cut And
    Inapproximability of Maximum Biclique Problems, Minimum k-Cut and Densest At-Least-k-Subgraph from the Small Set Expansion Hypothesis∗ Pasin Manurangsi† UC Berkeley May 11, 2017 Abstract The Small Set Expansion Hypothesis is a conjecture which roughly states that it is NP-hard to distinguish between a graph with a small subset of vertices whose (edge) expansion is almost zero and one in which all small subsets of vertices have expansion almost one. In this work, we prove conditional inapproximability results for the following graph problems based on this hypothesis: • Maximum Edge Biclique: given a bipartite graph G, find a complete bipartite subgraph of G that contains maximum number of edges. We show that, assuming the Small Set Expansion Hypothesis and that NP * BPP, no polynomial time algorithm approximates Maximum Edge Biclique to within a factor of n1−ε of the optimum for every constant ε> 0. • Maximum Balanced Biclique: given a bipartite graph G, find a balanced complete bipartite subgraph of G that contains maximum number of vertices. Similar to Maximum Edge Biclique, we prove that this problem is inapproximable in polynomial time to within a factor of n1−ε of the optimum for every constant ε> 0, assuming the Small Set Expansion Hypothesis and that NP * BPP. • Minimum k-Cut: given a weighted graph G, find a set of edges with minimum total weight whose removal partitions the graph into (at least) k connected components. For this problem, we prove that it is NP-hard to approximate it to within (2 − ε) factor of the optimum for every constant ε> 0, assuming the Small Set Expansion Hypothesis.
    [Show full text]
  • Planted Clique Detection Below the Noise Floor Using Low-Rank Sparse Pca
    PLANTED CLIQUE DETECTION BELOW THE NOISE FLOOR USING LOW-RANK SPARSE PCA Alexis B. Cook1 and Benjamin A. Miller2 1Department of Applied Mathematics, Brown University, Providence, RI 02906 alexis [email protected] 2MIT Lincoln Laboratory, Lexington, MA 02420 [email protected] ABSTRACT We study approaches to solving the planted clique problem that are based on an analysis of the spectral properties of the graph’s Detection of clusters and communities in graphs is useful in a wide modularity matrix. Nadakuditi proved in [8] the presence of a phase range of applications. In this paper we investigate the problem of transition between a regime in which the principal component of detecting a clique embedded in a random graph. Recent results have the modularity matrix can clearly identify the location of the clique demonstrated a sharp detectability threshold for a simple algorithm vertices and a regime in which it cannot. Through understanding the based on principal component analysis (PCA). Sparse PCA of the breakdown point of this simple spectral method for clique detection graph’s modularity matrix can successfully discover clique locations in high dimensional settings, we hope to motivate new algorithms to where PCA-based detection methods fail. In this paper, we demon- detect cliques smaller than previously thought possible. strate that applying sparse PCA to low-rank approximations of the Sparse principal component analysis (SPCA) has recently been modularity matrix is a viable solution to the planted clique problem shown to enable detection of dense subgraphs that cannot be detected that enables detection of small planted cliques in graphs where run- in the first principal component [7, 9].
    [Show full text]