Deep learning and the information bottleneck principle
Naftali Tishby,Noga Zaslavsky +1 more
- pp 1-5
TLDR
It is argued that both the optimal architecture, number of layers and features/connections at each layer, are related to the bifurcation points of the information bottleneck tradeoff, namely, relevant compression of the input layer with respect to the output layer.Abstract:
Deep Neural Networks (DNNs) are analyzed via the theoretical framework of the information bottleneck (IB) principle. We first show that any DNN can be quantified by the mutual information between the layers and the input and output variables. Using this representation we can calculate the optimal information theoretic limits of the DNN and obtain finite sample generalization bounds. The advantage of getting closer to the theoretical limit is quantifiable both by the generalization bound and by the network's simplicity. We argue that both the optimal architecture, number of layers and features/connections at each layer, are related to the bifurcation points of the information bottleneck tradeoff, namely, relevant compression of the input layer with respect to the output layer. The hierarchical representations at the layered network naturally correspond to the structural phase transitions along the information curve. We believe that this new insight can lead to new optimality bounds and deep learning algorithms.read more
Citations
More filters
Journal ArticleDOI
A Functional Contextual Account of Background Knowledge in Categorization: Implications for Artificial General Intelligence and Cognitive Accounts of General Knowledge
TL;DR: Relational Frame Theory (RFT) as discussed by the authors is a non-similarity-based post-Skinnerian theory of human language which is rooted in a philosophical world view called functional contextualism.
Journal ArticleDOI
Multi-feature deep information bottleneck network for breast cancer classification in contrast enhanced spectral mammography
TL;DR: In this article , a multi-feature deep information bottleneck (MDIB) was proposed for breast cancer classification in contrast enhanced spectral mammography (CESM) images, which incorporated an information bottleneck based module to learn the prominent representation that provide concise input while informative for the classification.
Proceedings ArticleDOI
Adaptive Label Smoothing for Classifier-based Mutual Information Neural Estimation
TL;DR: In this paper, the authors proposed a novel scheme that smooths the label adaptively according to how extreme the probability estimates of the classifier are, and the resulting MI estimate is unbiased under a mild assumption on the model.
Proceedings Article
A Framework for Transformation Network Training in Coordination with Semi-trusted Cloud Provider for Privacy-Preserving Deep Neural Networks
TL;DR: Wang et al. as mentioned in this paper proposed a framework for transformation network training in coordination with a semi-trusted cloud provider for privacy-preserving DNNs, where a user trained a transformation network using a model that a cloud provider has for transforming plain images into visually protected ones.
Proceedings Article
Improving Unsupervised Domain Adaptation with Variational Information Bottleneck.
TL;DR: In this paper, a variational bottleneck domain adaptation (VBDA) method is proposed to improve feature transferability by explicitly enforcing the feature extractor to ignore the task-irrelevant factors and focus on the information that is essential to the task of interest for both source and target domains.
References
More filters
Proceedings Article
ImageNet Classification with Deep Convolutional Neural Networks
TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Book
Elements of information theory
Thomas M. Cover,Joy A. Thomas +1 more
TL;DR: The author examines the role of entropy, inequality, and randomness in the design of codes and the construction of codes in the rapidly changing environment.
Journal ArticleDOI
Reducing the Dimensionality of Data with Neural Networks
TL;DR: In this article, an effective way of initializing the weights that allows deep autoencoder networks to learn low-dimensional codes that work much better than principal components analysis as a tool to reduce the dimensionality of data is described.
Journal ArticleDOI
Representation Learning: A Review and New Perspectives
TL;DR: Recent work in the area of unsupervised feature learning and deep learning is reviewed, covering advances in probabilistic models, autoencoders, manifold learning, and deep networks.
Book
Learning Deep Architectures for AI
TL;DR: The motivations and principles regarding learning algorithms for deep architectures, in particular those exploiting as building blocks unsupervised learning of single-layer modelssuch as Restricted Boltzmann Machines, used to construct deeper models such as Deep Belief Networks are discussed.