Information Bottleneck: Theory and Applications in Deep Learning

被引：9

作者：

Geiger, Bernhard C. ^{[1
]}

Kubin, Gernot ^{[2
]}

机构：

[1] Know Ctr GmbH, Inffeldgasse 13-6, A-8010 Graz, Austria

[2] Graz Univ Technol, Signal Proc & Speech Commun Lab, Inffeldgasse 16c, A-8010 Graz, Austria

来源：

基金：

欧盟地平线“2020”;

关键词：

information bottleneck; deep learning; neural networks;

D O I：

10.3390/e22121408

中图分类号：

O4 [物理学];

学科分类号：

0702 ;

摘要：

引用

页数：4

共 22 条

[11] The information bottleneck problem and its applications in machine learning
Goldfeld Z.
Polyanskiy Y.
[J]. IEEE Journal on Selected Areas in Information Theory, 2020, 1 (01): : 19 - 38
[12] Hendrycks Dan, 2019, INT C LEARN REPR
[13] Convergence Behavior of DNNs with Mutual-Information-Based Regularization
Jonsson, Hlynur
Cherubini, Giovanni
Eleftheriou, Evangelos
[J]. ENTROPY, 2020, 22 (07)
[14] Kolchinsky A., 2019, P INT C LEARN REPR I
[15] Nonlinear Information Bottleneck
Kolchinsky, Artemy
Tracey, Brendan D.
Wolpert, David H.
[J]. ENTROPY, 2019, 21 (12)
[16] Gaussian Mean Field Regularizes by Limiting Learned Information
Kunze, Julius
Kirsch, Louis
Ritter, Hippolyt
Barber, David
[J]. ENTROPY, 2019, 21 (08)
[17] Shwartz-Ziv Ravid, 2017, OPENING BLACK BOX DE
[18] Pareto-Optimal Data Compression for Binary Classification Tasks
Tegmark, Max
Wu, Tailin
[J]. ENTROPY, 2020, 22 (01) : 7
[19] Markov Information Bottleneck to Improve Information Flow in Stochastic Neural Networks
Thanh Tang Nguyen
Choi, Jaesik
[J]. ENTROPY, 2019, 21 (10)
[20] Tishby N., 1999, P 37 ANN ALL C COMM, P368, DOI DOI 10.48550/ARXIV.PHYSICS/0004057