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
Neural Joint Entropy Estimation
TL;DR: In this article , the authors introduce a family of estimators for related information-theoretic measures, such as conditional entropy and mutual information, and apply the proposed estimators to conditional MI estimation, as well as focus on independence testing tasks.
Journal ArticleDOI
Photo-Realistic Out-of-domain GAN inversion via Invertibility Decomposition
TL;DR: In this paper , a spatial alignment module is proposed to align the generated features to the input geome-try and reduce the reconstruction error in the OOD regions, which can be more distinguishable and can be precisely predicted.
Posted Content
Pro-KD: Progressive Distillation by Following the Footsteps of the Teacher.
TL;DR: Pro-KD as mentioned in this paper defines a smoother training path for the student by following the training footprints of the teacher instead of solely relying on distilling from a single mature fully-trained teacher.
Posted Content
Fundamental Limits of Online Learning: An Entropic-Innovations Viewpoint
Song Fang,Quanyan Zhu +1 more
TL;DR: This paper examines the fundamental performance limitations of online machine learning, by viewing the online learning problem as a prediction problem with causal side information, and derives generic lower bounds on the prediction errors as well as the conditions to achieve the bounds.
Proceedings ArticleDOI
Compressed Data Sharing Based On Information Bottleneck Model
TL;DR: This paper considers privacy-preserving compressed image sharing, where the goal is to release compressed data whilst satisfying some privacy/secrecy constraints yet ensuring image reconstruction with a defined fidelity, using a machine learning framework based on an information bottleneck with a shared secret key for authorized users.
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.