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
Attention Augmented Face Morph Detection
TL;DR: In this paper , the authors proposed an end-to-end attention-based deep morph detector which assimilates the most discriminative wavelet sub-bands of a given image which are obtained by a group sparsity representation learning scheme.
Journal ArticleDOI
Conditional information gain networks as sparse mixture of experts
Ufuk Can Bicici,Lale Akarun +1 more
TL;DR: In this article, a sparse mixture of experts (sMoE-CIGNs) model is proposed, which groups semantically similar samples in the leaves, enabling simpler classifiers to focus on differentiating between similar classes.
Proceedings ArticleDOI
On The Energy Statistics of Feature Maps in Pruning of Neural Networks with Skip-Connections
TL;DR: A new structured pruning framework for compressing Deep Neural Networks with skip-connections is proposed, based on measuring the statistical dependency of hidden layers and predicted outputs, with competitive performance to state-of-the-art methods.
Proceedings ArticleDOI
Feature Map Alignment: Towards Efficient Design of Mixed-precision Quantization Scheme
Yukun Bao,Yuhui Xu,Hongkai Xiong +2 more
TL;DR: This paper proposes a novel post-training quantization approach which derives a flexible bitwidth scheme and achieves impressive results for mainstream neural networks.
Posted Content
Deep Learning for Functional Data Analysis with Adaptive Basis Layers
TL;DR: In this article, the hidden units are each basis functions themselves implemented as a micro neural network, which learns to apply parsimonious dimension reduction to functional inputs that focuses only on information relevant to the target rather than irrelevant variation in the input function.
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.