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
Posted Content
Critical Slowing Down Near Topological Transitions in Rate-Distortion Problems
TL;DR: In this article, the authors study the convergence time of the Arimoto-Blahut alternating projection algorithm near critical points, both for the rate-distortion and information bottleneck settings.
Journal ArticleDOI
Quadratic Privacy-Signaling Games and the MMSE Information Bottleneck Problem for Gaussian Sources
TL;DR: In this paper , the authors investigated a privacy-signaling game in which a sender with privacy concerns observes a pair of correlated random vectors which are modeled as jointly Gaussian, and the objective of the receiver is to accurately estimate both of the random vectors.
Journal ArticleDOI
Heuristic Attention Representation Learning for Self-Supervised Pretraining
TL;DR: HARL framework adopts prior visual object-level attention by generating a heuristic mask proposal for each training image and maximizes the abstract object- level embedding on vector space instead of whole image representation from previous works, and outperforms existing self-supervised baselines on several downstream tasks.
Proceedings ArticleDOI
Interpretability with full complexity by constraining feature information
TL;DR: The authors use the distributed information bottleneck to find optimal compressions of each feature that maximally preserve information about the output of a model, which can provide rich opportunities for interpretation, particularly in problems with many features and complex feature interactions.
Proceedings ArticleDOI
Contrastive Collaborative Filtering for Cold-Start Item Recommendation
Zhihui Zhou,Lili Zhang,Ning Yang +2 more
TL;DR: Zhang et al. as discussed by the authors proposed a contrastive collaborative filtering (CF) framework, consisting of a content CF module and a co-occurrence CF module to generate the content-based collaborative embedding and the cooccurrence collaborative embeddings for a training item, respectively.
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.