Information theoretic limits of learning a sparse rule

被引:0
|
作者
Luneau, Clement [1 ]
Macris, Nicolas [1 ]
Barbier, Jean [2 ]
机构
[1] Ecole Polytech Fed Lausanne, Lausanne, Switzerland
[2] Abdus Salaam Int Ctr Theoret Phys, Trieste, Italy
关键词
MUTUAL INFORMATION; TIGHT BOUNDS; SHARP BOUNDS; CAPACITY; ERROR;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We consider generalized linear models in regimes where the number of nonzero components of the signal and accessible data points are sublinear with respect to the size of the signal. We prove a variational formula for the asymptotic mutual information per sample when the system size grows to infinity. This result allows us to derive an expression for the minimum mean-square error (MMSE) of the Bayesian estimator when the signal entries have a discrete distribution with finite support. We find that, for such signals and suitable vanishing scalings of the sparsity and sampling rate, the MMSE is nonincreasing piecewise constant. In specific instances the MMSE even displays an all-or-nothing phase transition, that is, the MMSE sharply jumps from its maximum value to zero at a critical sampling rate. The all-or-nothing phenomenon has previously been shown to occur in high-dimensional linear regression. Our analysis goes beyond the linear case and applies to learning the weights of a perceptron with general activation function in a teacher-student scenario. In particular, we discuss an all-or-nothing phenomenon for the generalization error with a sublinear set of training examples.
引用
收藏
页数:12
相关论文
共 50 条
  • [21] Information-theoretic Limits of Subspace Clustering
    Ahn, Kwangjun
    Lee, Kangwook
    Suh, Changho
    2017 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2017, : 2473 - 2477
  • [22] Information-Theoretic Limits of Matrix Completion
    Riegler, Erwin
    Stotz, David
    Bolcskei, Helmut
    2015 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2015, : 1836 - 1840
  • [23] Computational Techniques for Investigating Information Theoretic Limits of Information Systems
    Tian, Chao
    Plank, James S.
    Hurst, Brent
    Zhou, Ruida
    INFORMATION, 2021, 12 (02) : 1 - 16
  • [24] Side Information in Recovering a Single Community: Information Theoretic Limits
    Saad, Hussein
    Nosratinia, Aria
    2018 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2018, : 2107 - 2111
  • [25] Information-Theoretic Bounds for Adaptive Sparse Recovery
    Aksoylar, Cem
    Saligrama, Venkatesh
    2014 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2014, : 1311 - 1315
  • [26] Information Theoretic Bounds for Sparse Reconstruction in Random Noise
    Chen, Junjie
    Zhu, Fangqi
    Liang, Qilian
    IEEE ACCESS, 2019, 7 (102304-102312) : 102304 - 102312
  • [27] AN INFORMATION THEORETIC APPROACH TO RULE INDUCTION FROM DATABASES
    SMYTH, P
    GOODMAN, RM
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 1992, 4 (04) : 301 - 316
  • [28] The information and wave-theoretic limits of analog beamforming
    Mezghani, Amine
    Heath, Robert W., Jr.
    2018 INFORMATION THEORY AND APPLICATIONS WORKSHOP (ITA), 2018,
  • [29] Information-theoretic limits for the matrix tensor product
    Reeves G.
    IEEE Journal on Selected Areas in Information Theory, 2020, 1 (03): : 777 - 798
  • [30] Information-Theoretic Limits of Dense Underwater Networks
    Shin, Won-Yong
    Lucani, Daniel E.
    Medard, Muriel
    Stojanovic, Milica
    Tarokh, Vahid
    2011 CONFERENCE RECORD OF THE FORTY-FIFTH ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS (ASILOMAR), 2011, : 1835 - 1839