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
Dissertation
Parametric Information Bottleneck to \\Optimize Stochastic Neural Networks
Thanh Tang Nguyen,Jaesik Choi +1 more
TL;DR: This paper proposes Parametric Information Bottleneck (PIB) for a neural network by utilizing (only) its model parameters explicitly to approximate the compression and the relevance and shows that PIBs improve the generalization of neural networks in classification tasks, and push the representation of Neural networks closer to the optimal information-theoretical representation in a faster manner.
Task-Oriented Semantic Communication with Semantic Reconstruction: An Extended Rate-Distortion Theory Based Scheme
TL;DR: This paper formulate the TOSC-SR scheme as a rate-distortion optimization problem, where a novel semantic distortion measurement is defined by mutual information of source, the semantic-reconstructed images, and task labels, pairwise and derive an analytic solution for the formulated problem.
Posted Content
Analyzing Data Selection Techniques with Tools from the Theory of Information Losses
Brandon Foggo,Nanpeng Yu +1 more
TL;DR: This paper presents and illustrates some new tools for rigorously analyzing training data selection methods and proves that two methods, Facility Location Selection and Transductive Experimental Design, reduce these losses.
Book ChapterDOI
Detecting Defects in Materials Using Deep Convolutional Neural Networks
TL;DR: This paper proposes representing and detecting manufacturing defects at the micrometre scale using deep convolutional neural networks, and investigates a variety of design parameters pertaining to data preprocessing and network architecture.
Journal ArticleDOI
Interpreting rate-distortion of variational autoencoder and using model uncertainty for anomaly detection
TL;DR: This work revisits VAE from the perspective of information theory to provide some theoretical foundations on using the reconstruction error, and incorporates a practical model uncertainty measure into the metric to enhance the effectiveness of detecting anomalies.
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.