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
Journal ArticleDOI
Texture aware autoencoder pre-training and pairwise learning refinement for improved iris recognition
TL;DR: In this article , a texture-aware end-to-end trainable iris recognition system was proposed, where a better autoencoding framework with a data relation loss between Gram matrix representations of input and reconstructed images was proposed.
Posted Content
Pointly-supervised 3D Scene Parsing with Viewpoint Bottleneck.
TL;DR: In this article, a self-supervised 3D representation learning framework named viewpoint bottleneck is proposed to leverage numerous unlabeled points in 3D point clouds, which optimizes a mutual-information based objective, which is applied on point clouds under different viewpoints.
Book ChapterDOI
Whatever Happened to the Logic of Discovery? From Transparent Logic to Alien Reasoning
Journal ArticleDOI
GINT: A Generative Interpretability method via perturbation in the latent space
TL;DR: In this article , a generative model is adopted to generate perturbations in the latent space instead of randomly perturbing the feature space, which is called Generative InTerpretability (GINT) method.
Posted Content
Quantization-Based Regularization for Autoencoders.
TL;DR: This paper introduces a quantization-based regularizer in the bottleneck stage of autoencoder models to learn meaningful latent representations and shows that the proposed regularization method results in improved latent representations for both supervised learning and clustering downstream tasks when compared to autoencoders using other bottleneck structures.
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.