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
RD-DPP: Rate-Distortion Theory Meets Determinantal Point Process to Diversify Learning Data Samples
TL;DR: In this article , the authors proposed a new way of measuring task-oriented diversity based on the Rate-Distortion (RD) theory, appropriate for multi-level classification and established a fundamental relationship between DPP and RD theory to evaluate the diversity gain of data samples.
Posted Content
Task-Discriminative Domain Alignment for Unsupervised Domain Adaptation
TL;DR: In this article, a task-driven domain alignment discriminator is proposed to encourage the shared features as task-specific and domain invariant, and prompt the task model to be data structure preserving, guiding its decision boundaries through the low density data regions.
Dissertation
Extracting Compact Knowledge From Massive Data
TL;DR: A selection of photos from the 2016/17 USGS report on quantitative hazard assessments of earthquake-triggered landsliding and liquefaction.
Journal ArticleDOI
Representation Learning for Dynamic Functional Connectivities via Variational Dynamic Graph Latent Variable Models
Yi Huang,Zhu Liang Yu +1 more
TL;DR: This work introduces a dynamic graph as the latent variable and develops a Variational Dynamic Graph Latent Variable Model (VDGLVM), a representation learning model based on the variational information bottleneck framework that provides guaranteed behavior-decoding performance and improves LVMs by associating the inferred latent dynamics with probable DFC.
Proceedings ArticleDOI
Towards Universal Cross-Domain Recommendation
TL;DR: UniCDR as discussed by the authors is a unified framework for cross-domain recommendation, which can universally model different CDR scenarios by transferring the domain-shared information, and it can transfer rich user-item interaction information from related source domains to boost recommendation quality of target domains.
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.