Suppr超能文献

半定松弛中的相变。

Phase transitions in semidefinite relaxations.

作者信息

Javanmard Adel, Montanari Andrea, Ricci-Tersenghi Federico

机构信息

Marshall School of Business, University of Southern California, Los Angeles, CA 90089;

Department of Statistics, Stanford University, Stanford, CA 94305; Department of Electrical Engineering, Stanford University, Stanford, CA 94305;

出版信息

Proc Natl Acad Sci U S A. 2016 Apr 19;113(16):E2218-23. doi: 10.1073/pnas.1523097113. Epub 2016 Mar 21.

Abstract

Statistical inference problems arising within signal processing, data mining, and machine learning naturally give rise to hard combinatorial optimization problems. These problems become intractable when the dimensionality of the data is large, as is often the case for modern datasets. A popular idea is to construct convex relaxations of these combinatorial problems, which can be solved efficiently for large-scale datasets. Semidefinite programming (SDP) relaxations are among the most powerful methods in this family and are surprisingly well suited for a broad range of problems where data take the form of matrices or graphs. It has been observed several times that when the statistical noise is small enough, SDP relaxations correctly detect the underlying combinatorial structures. In this paper we develop asymptotic predictions for several detection thresholds, as well as for the estimation error above these thresholds. We study some classical SDP relaxations for statistical problems motivated by graph synchronization and community detection in networks. We map these optimization problems to statistical mechanics models with vector spins and use nonrigorous techniques from statistical mechanics to characterize the corresponding phase transitions. Our results clarify the effectiveness of SDP relaxations in solving high-dimensional statistical problems.

摘要

信号处理、数据挖掘和机器学习中出现的统计推断问题自然会引发棘手的组合优化问题。当数据维度很大时,这些问题就变得难以处理,现代数据集往往就是这种情况。一个流行的想法是构建这些组合问题的凸松弛,对于大规模数据集,这种凸松弛可以有效求解。半定规划(SDP)松弛是这类方法中最强大的方法之一,并且令人惊讶地非常适合数据采用矩阵或图形式的广泛问题。多次观察到,当统计噪声足够小时,SDP松弛能正确检测到潜在的组合结构。在本文中,我们针对几个检测阈值以及这些阈值之上的估计误差给出渐近预测。我们研究一些受网络中的图同步和社区检测启发的统计问题的经典SDP松弛。我们将这些优化问题映射到具有向量自旋的统计力学模型,并使用统计力学中的非严格技术来刻画相应的相变。我们的结果阐明了SDP松弛在解决高维统计问题中的有效性。

相似文献

1
Phase transitions in semidefinite relaxations.
Proc Natl Acad Sci U S A. 2016 Apr 19;113(16):E2218-23. doi: 10.1073/pnas.1523097113. Epub 2016 Mar 21.
2
Worst case linear discriminant analysis as scalable semidefinite feasibility problems.
IEEE Trans Image Process. 2015 Aug;24(8):2382-92. doi: 10.1109/TIP.2015.2401511. Epub 2015 Feb 6.
3
Facial reduction for symmetry reduced semidefinite and doubly nonnegative programs.
Math Program. 2023;200(1):475-529. doi: 10.1007/s10107-022-01890-9. Epub 2022 Sep 27.
4
Solving large-scale general phase retrieval problems via a sequence of convex relaxations.
J Opt Soc Am A Opt Image Sci Vis. 2018 Aug 1;35(8):1410-1419. doi: 10.1364/JOSAA.35.001410.
5
Certifiably Optimal Outlier-Robust Geometric Perception: Semidefinite Relaxations and Scalable Global Optimization.
IEEE Trans Pattern Anal Mach Intell. 2023 Mar;45(3):2816-2834. doi: 10.1109/TPAMI.2022.3179463. Epub 2023 Feb 3.
6
A computational study of exact subgraph based SDP bounds for Max-Cut, stable set and coloring.
Math Program. 2020;183(1):283-308. doi: 10.1007/s10107-020-01512-2. Epub 2020 May 25.
7
Optimal bounds with semidefinite programming: An application to stress-driven shear flows.
Phys Rev E. 2016 Apr;93:043308. doi: 10.1103/PhysRevE.93.043308. Epub 2016 Apr 8.
8
Observed universality of phase transitions in high-dimensional geometry, with implications for modern data analysis and signal processing.
Philos Trans A Math Phys Eng Sci. 2009 Nov 13;367(1906):4273-93. doi: 10.1098/rsta.2009.0152.
9
Solving condensed-matter ground-state problems by semidefinite relaxations.
Phys Rev Lett. 2012 May 18;108(20):200404. doi: 10.1103/PhysRevLett.108.200404. Epub 2012 May 17.
10
An SDP-based approach for computing the stability number of a graph.
Math Methods Oper Res (Heidelb). 2022;95(1):141-161. doi: 10.1007/s00186-022-00773-1. Epub 2022 Mar 12.

引用本文的文献

2
Approximate message passing from random initialization with applications to synchronization.
Proc Natl Acad Sci U S A. 2023 Aug;120(31):e2302930120. doi: 10.1073/pnas.2302930120. Epub 2023 Jul 25.
3
Distributed Certifiably Correct Pose-Graph Optimization.
IEEE Trans Robot. 2021 Dec;37(6):2137-2156. doi: 10.1109/tro.2021.3072346. Epub 2021 May 7.
4
ENTRYWISE EIGENVECTOR ANALYSIS OF RANDOM MATRICES WITH LOW EXPECTED RANK.
Ann Stat. 2020 Jun;48(3):1452-1474. doi: 10.1214/19-aos1854. Epub 2020 Jul 17.
5
No Statistical-Computational Gap in Spiked Matrix Models with Generative Network Priors.
Entropy (Basel). 2021 Jan 16;23(1):115. doi: 10.3390/e23010115.
6
Kernel k-Groups via Hartigan's Method.
IEEE Trans Pattern Anal Mach Intell. 2021 Dec;43(12):4411-4425. doi: 10.1109/TPAMI.2020.2998120. Epub 2021 Nov 3.
7
Universal Completability, Least Eigenvalue Frameworks, and Vector Colorings.
Discrete Comput Geom. 2017;58(2):265-292. doi: 10.1007/s00454-017-9899-2. Epub 2017 Jun 19.
8
Inference on graphs via semidefinite programming.
Proc Natl Acad Sci U S A. 2016 Apr 19;113(16):4238-9. doi: 10.1073/pnas.1603405113. Epub 2016 Apr 8.

本文引用的文献

1
Spectral redemption in clustering sparse networks.
Proc Natl Acad Sci U S A. 2013 Dec 24;110(52):20935-40. doi: 10.1073/pnas.1312486110. Epub 2013 Nov 25.
3
Asymptotic analysis of the stochastic block model for modular networks and its algorithmic applications.
Phys Rev E Stat Nonlin Soft Matter Phys. 2011 Dec;84(6 Pt 2):066106. doi: 10.1103/PhysRevE.84.066106. Epub 2011 Dec 12.
4
Angular Synchronization by Eigenvectors and Semidefinite Programming.
Appl Comput Harmon Anal. 2011 Jan 30;30(1):20-36. doi: 10.1016/j.acha.2010.02.001.
5
Message-passing algorithms for compressed sensing.
Proc Natl Acad Sci U S A. 2009 Nov 10;106(45):18914-9. doi: 10.1073/pnas.0909892106. Epub 2009 Oct 26.
6
Neighborliness of randomly projected simplices in high dimensions.
Proc Natl Acad Sci U S A. 2005 Jul 5;102(27):9452-7. doi: 10.1073/pnas.0502258102. Epub 2005 Jun 22.
7
Community structure in social and biological networks.
Proc Natl Acad Sci U S A. 2002 Jun 11;99(12):7821-6. doi: 10.1073/pnas.122653799.
8
Bayesian Model Selection and Model Averaging.
J Math Psychol. 2000 Mar;44(1):92-107. doi: 10.1006/jmps.1999.1278.
9
Clustering gene expression patterns.
J Comput Biol. 1999 Fall-Winter;6(3-4):281-97. doi: 10.1089/106652799318274.

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验