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 ContentDOI
Expansion and contraction of resource allocation in sensory bottlenecks
TL;DR: In this article, a sensory bottleneck is introduced to impose constraints on resource allocation and derive the optimal neural allocation, inducing either expansion or contraction of the cortical representation, based on a linear second-order models that maximize information transmission through decorrelation.
Journal ArticleDOI
Towards Explanation for Unsupervised Graph-Level Representation Learning
TL;DR: By only keeping frequent phrases in the training examples, state-of-the-art models perform similarly compared to the variant trained with full data, suggesting they exploit these spurious correlations to solve the task.
Journal ArticleDOI
Bounds on mutual information of mixture data for classification tasks
Amit Ashok,Lorenz Böllinger +1 more
TL;DR: In this paper , a variational upper bound, a lower bound, and three approximations, all employing pair-wise divergences between mixture components, were proposed. And they were compared with Monte Carlo stochastic sampling and bounds derived from entropy bounds.
Posted Content
Reducing Information Bottleneck for Weakly Supervised Semantic Segmentation.
TL;DR: In this article, the authors propose a method to reduce the information bottleneck by removing the last activation function of the last layer of a deep neural network, which is activated by the sigmoid or softmax activation functions, and as a result, only a subset of the task-relevant information is passed on to the output.
Journal ArticleDOI
Vision Learners Meet Web Image-Text Pairs
TL;DR: MUG as mentioned in this paper is a self-supervised pre-training method that learns from scalable web sourced image-text paired data and achieves promising scaling properties, achieving state-of-the-art transfer performance.
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.