scispace - formally typeset
Open AccessProceedings ArticleDOI

Deep learning and the information bottleneck principle

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
Book ChapterDOI

An Application of Data Compression Models to Handwritten Digit Classification

TL;DR: This paper addresses handwritten digit classification as a special problem of data compression modeling and shows that it is possible to attain a classification accuracy consistently above 99.3% on the MNIST dataset, using classifiers trained in less than one hour on a common laptop.
Journal ArticleDOI

On the Design Fundamentals of Diffusion Models: A Survey

TL;DR: In this article , a comprehensive and coherent review on component-wise design choices in diffusion models is provided, where the authors organize diffusion models according to their three key components, namely the forward process, the reverse process, and the sampling procedure.
Posted Content

Robust Predictable Control

TL;DR: In this article, the authors combine information bottlenecks, model-based RL and bits-back coding into a simple and theoretically-justified algorithm to jointly optimize a latent space model and policy to be self-consistent, such that the policy avoids states where the model is inaccurate.
Book ChapterDOI

Anomaly Detection via Few-Shot Learning on Normality

Shin Ando, +1 more
TL;DR: This paper addresses deep anomaly detection aimed at embedding ‘ normal’ classes to individually close but mutually distant proximities by introducing a problem setting where a limited number of labeled examples from each ‘normal’ class is available for training.
Journal ArticleDOI

Higher-order mutual information reveals synergistic sub-networks for multi-neuron importance

TL;DR: In this article , a multivariate expansion of the mutual information (O-information) was used to identify which groups of neurons contain synergistic or redundant information using a multi-layer network.
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

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.