Information Bottleneck Theory and Applications in Deep Learning

The celebrated information bottleneck (IB) principle of Tishby et al. has recently enjoyed renewed attention due to its application in the area of deep learning. This collection investigates the IB principle in this new context. The individual chapters in this collection: • provide novel insights in...

Full description

Bibliographic Details
Main Author: Geiger, Bernhard
Other Authors: Kubin, Gernot
Format: eBook
Language:English
Published: Basel, Switzerland MDPI - Multidisciplinary Digital Publishing Institute 2021
Subjects:
Online Access:
Collection: Directory of Open Access Books - Collection details see MPG.ReNa
LEADER 03015nma a2200661 u 4500
001 EB002049647
003 EBX01000000000000001193313
005 00000000000000.0
007 cr|||||||||||||||||||||
008 220822 ||| eng
020 |a books978-3-0365-0803-0 
020 |a 9783036508023 
020 |a 9783036508030 
100 1 |a Geiger, Bernhard 
245 0 0 |a Information Bottleneck  |h Elektronische Ressource  |b Theory and Applications in Deep Learning 
260 |a Basel, Switzerland  |b MDPI - Multidisciplinary Digital Publishing Institute  |c 2021 
300 |a 1 electronic resource (274 p.) 
653 |a machine learning 
653 |a information theory 
653 |a conspicuous subset 
653 |a hand crafted priors 
653 |a deep networks 
653 |a semi-supervised classification 
653 |a representation learning 
653 |a deep learning 
653 |a neural networks 
653 |a learnable priors 
653 |a variational inference 
653 |a compression 
653 |a deep neural networks 
653 |a learnability 
653 |a latent space representation 
653 |a information bottleneck principle 
653 |a classifier 
653 |a Information technology industries / bicssc 
653 |a information bottleneck 
653 |a stochastic neural networks 
653 |a ensemble 
653 |a classification 
653 |a decision tree 
653 |a regularization methods 
653 |a optimization 
653 |a mutual information 
653 |a information 
653 |a regularization 
653 |a bottleneck 
700 1 |a Kubin, Gernot 
700 1 |a Geiger, Bernhard 
700 1 |a Kubin, Gernot 
041 0 7 |a eng  |2 ISO 639-2 
989 |b DOAB  |a Directory of Open Access Books 
500 |a Creative Commons (cc), https://creativecommons.org/licenses/by/4.0/ 
028 5 0 |a 10.3390/books978-3-0365-0803-0 
856 4 2 |u https://directory.doabooks.org/handle/20.500.12854/76429  |z DOAB: description of the publication 
856 4 0 |u https://www.mdpi.com/books/pdfview/book/3864  |7 0  |x Verlag  |3 Volltext 
082 0 |a 000 
082 0 |a 600 
520 |a The celebrated information bottleneck (IB) principle of Tishby et al. has recently enjoyed renewed attention due to its application in the area of deep learning. This collection investigates the IB principle in this new context. The individual chapters in this collection: • provide novel insights into the functional properties of the IB; • discuss the IB principle (and its derivates) as an objective for training multi-layer machine learning structures such as neural networks and decision trees; and • offer a new perspective on neural network learning via the lens of the IB framework. Our collection thus contributes to a better understanding of the IB principle specifically for deep learning and, more generally, of information-theoretic cost functions in machine learning. This paves the way toward explainable artificial intelligence.