Computational phase transitions in sparse planted problems?

In recent times the cavity method, a statistical physics-inspired heuristic, has been successful in conjecturing computational thresholds that have been rigorously con rmed — such as for community detection in the sparse regime of the stochastic block model. Inspired by this, we investigate the predictions made by the cavity method for the algorithmic problems of detecting and recovering a planted signal in a general model of sparse random graphs. The model we study generalizes the well-understood case of the stochastic block model, the less well understood case of random constraint satisfaction problems with planted assignments, as well as “semi-supervised” variants of these models. Our results include: (i) a conjecture about a precise criterion for when the problems of detection and recovery should be algorithmically tractable arising from a heuristic analysis of when a particular xed point of the belief propagation algorithm is stable; (ii) a rigorous polynomial-time algorithm for the problem of detection: distinguishing a graph with a planted signal from one without; (iii) a rigorous polynomial-time algorithm for the problem of recovery: outputting a vector that correlates with the planted signal signi cantly better than a random guess would. The rigorous algorithms are based on the spectra of matrices that arise as the derivatives of the belief propagation update rule. An interesting unanswered question raised is that of obtaining evidence of computational hardness for convex relaxations whenever hardness is predicted by the cavity method. ∗EECS Department, University of California Berkeley. sliu18@berkeley.edu. This research was supported in part by a Google Faculty Award and donations from the Ethereum Foundation and the Interchain Foundation. †EECS Department, University of California at Berkeley. sidhanthm@berkeley.edu. Supported by Google PhD Fellowship. ‡EECS Department, University of California Berkeley. prasad@cs.berkeley.edu. Supported by NSF grant CCF1718695. ar X iv :2 10 1. 10 88 2v 1 [ cs .D S] 2 6 Ja n 20 21

[1]  Charles Bordenave,et al.  A new proof of Friedman's second eigenvalue theorem and its extension to random lifts , 2015, Annales scientifiques de l'École normale supérieure.

[2]  Elchanan Mossel,et al.  Approximation Resistant Predicates from Pairwise Independence , 2008, 2008 23rd Annual IEEE Conference on Computational Complexity.

[3]  Tselil Schramm,et al.  Statistical Query Algorithms and Low-Degree Tests Are Almost Equivalent , 2020, COLT.

[4]  Ryan O'Donnell,et al.  Sum of squares lower bounds for refuting any CSP , 2017, STOC.

[5]  Prasad Raghavendra,et al.  Optimal algorithms and inapproximability results for every CSP? , 2008, STOC.

[6]  Laurent Massoulié,et al.  Community Detection in the Labelled Stochastic Block Model , 2012, ArXiv.

[7]  Elchanan Mossel,et al.  A Proof of the Block Model Threshold Conjecture , 2013, Combinatorica.

[8]  Subhash Khot,et al.  On the power of unique 2-prover 1-round games , 2002, Proceedings 17th IEEE Annual Conference on Computational Complexity.

[9]  B. Collins,et al.  Eigenvalues of random lifts and polynomials of random permutation matrices , 2018, Annals of Mathematics.

[10]  Oded Goldreich,et al.  Candidate One-Way Functions Based on Expander Graphs , 2011, Studies in Complexity and Cryptography.

[11]  Cristopher Moore,et al.  Phase transitions in semisupervised clustering of sparse networks , 2014, Physical review. E, Statistical, nonlinear, and soft matter physics.

[12]  Florent Krzakala,et al.  Statistical physics of inference: thresholds and algorithms , 2015, ArXiv.

[13]  Wasim Huleihel,et al.  Reducibility and Computational Lower Bounds for Problems with Planted Sparse Structure , 2018, COLT.

[14]  S. Kak.  Information, physics, and computation , 1996 .

[15]  Dmitriy Zhuk,et al.  A Proof of CSP Dichotomy Conjecture , 2017, 2017 IEEE 58th Annual Symposium on Foundations of Computer Science (FOCS).

[16]  David Steurer,et al.  Efficient Bayesian Estimation from Few Samples: Community Detection and Related Problems , 2017, 2017 IEEE 58th Annual Symposium on Foundations of Computer Science (FOCS).

[17]  Andrea Montanari,et al.  How well do local algorithms solve semidefinite programs? , 2016, STOC.

[18]  Laurent Massoulié,et al.  Community detection thresholds and the weak Ramanujan property , 2013, STOC.

[19]  Elchanan Mossel,et al.  Belief propagation, robust reconstruction and optimal recovery of block models , 2013, COLT.

[20]  Guy Bresler,et al.  Reducibility and Statistical-Computational Gaps from Secret Leakage , 2020, COLT.

[21]  Santosh S. Vempala,et al.  Statistical Algorithms and a Lower Bound for Detecting Planted Cliques , 2012, J. ACM.

[22]  Emmanuel Abbe,et al.  Community Detection in General Stochastic Block models: Fundamental Limits and Efficient Algorithms for Recovery , 2015, 2015 IEEE 56th Annual Symposium on Foundations of Computer Science.

[23]  Sidhanth Mohanty,et al.  The SDP value for random two-eigenvalue CSPs , 2019, STACS.

[24]  Joel Friedman,et al.  A proof of Alon's second eigenvalue conjecture and related problems , 2004, ArXiv.

[25]  Andrea Montanari,et al.  Contextual Stochastic Block Models , 2018, NeurIPS.

[26]  Andrea Montanari,et al.  The threshold for SDP-refutation of random regular NAE-3SAT , 2018, SODA.

[27]  Andrea Montanari,et al.  Estimating random variables from random sparse observations , 2007, Eur. Trans. Telecommun..

[28]  Ryan O'Donnell,et al.  How to Refute a Random CSP , 2015, 2015 IEEE 56th Annual Symposium on Foundations of Computer Science.

[29]  Guy Kindler,et al.  Optimal inapproximability results for MAX-CUT and other 2-variable CSPs? , 2004, 45th Annual IEEE Symposium on Foundations of Computer Science.

[30]  Prasad Raghavendra,et al.  Strongly refuting random CSPs below the spectral threshold , 2016, STOC.

[31]  Andrea Montanari,et al.  Estimation of low-rank matrices via approximate message passing , 2017, The Annals of Statistics.

[32]  Cristopher Moore,et al.  Asymptotic analysis of the stochastic block model for modular networks and its algorithmic applications , 2011, Physical review. E, Statistical, nonlinear, and soft matter physics.

[33]  Sidhanth Mohanty,et al.  Local Statistics, Semidefinite Programming, and Community Detection , 2019, SODA.

[34]  Laurent Massoulié,et al.  Non-backtracking Spectrum of Random Graphs: Community Detection and Non-regular Ramanujan Graphs , 2014, 2015 IEEE 56th Annual Symposium on Foundations of Computer Science.

[35]  Ryan O'Donnell,et al.  Explicit near-fully X-Ramanujan graphs , 2020, 2020 IEEE 61st Annual Symposium on Foundations of Computer Science (FOCS).

[36]  Florent Krzakala,et al.  Hiding Quiet Solutions in Random Constraint Satisfaction Problems , 2009, Physical review letters.