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
Task-Oriented Image Semantic Communication Based on Rate-Distortion Theory
TL;DR: In this paper , the authors propose a solution to solve the problem of the problem: this paper ] of "uniformity" and "uncertainty" of the solution.
Posted Content
On Equivariant and Invariant Learning of Object Landmark Representations
TL;DR: In this article, a simple and effective approach by combining instance-discriminative and spatially discriminative contrastive learning is developed to discover landmarks by modeling the shared geometric structure across instances.
Book ChapterDOI
Layer-Wise Entropy Analysis and Visualization of Neurons Activation
TL;DR: The entropy analysis is leveraged to study the neurons activation behavior of the fully connected layers of DNNs and it is found that for the well trained deep neural networks model, the entropy of the neuron activation pattern is monotonically reduced with the depth of the layers.
Journal ArticleDOI
Explainable Recommender with Geometric Information Bottleneck
TL;DR: In this article , a geometric prior learned from user-item interactions is incorporated into a variational network which infers latent factors from user and item interactions for both recommendation and explanation generation, which naturally inherit the global characteristics encoded in the prior knowledge.
Book ChapterDOI
Self-supervised Graph Learning with Segmented Graph Channels
TL;DR: By combining SGCs with data augmentation, SGL-SGC can generate views that vastly reduce the redundant information and design a feature-level weight-sensitive loss to jointly accelerate optimization and avoid the model falling into a local optimum.
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.