Learning Diverse and Discriminative Representations via the Principle of Maximal Coding Rate Reduction

Abstract

To learn intrinsic low-dimensional structures from high-dimensional data that most discriminate between classes, we propose the principle of Maximal Coding Rate Reduction (${\text{MCR}}^2$), an information-theoretic measure that maximizes the coding rate difference between the whole dataset and the sum of each individual class. We clarify its relationships with most existing frameworks such as cross-entropy, information bottleneck, information gain, contractive and contrastive learning, and provide theoretical guarantees for learning diverse and discriminative features. The coding rate can be accurately computed from finite samples of degenerate subspace-like distributions and can learn intrinsic representations in supervised, self-supervised, and unsupervised settings in a unified manner. Empirically, the representations learned using this principle alone are significantly more robust to label corruptions in classification than those using cross-entropy, and can lead to state-of-the-art results in clustering mixed data from self-learned invariant features.

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Momentum Contrast for Unsupervised Visual Representation Learning
• Joint Unsupervised Learning of Deep Representations and Image Clusters
• Oracle Based Active Set Algorithm for Scalable Elastic Net Subspace Clustering
• Deep Sparse Subspace Clustering
• A Rate-Distortion Framework for Explaining Neural Network Decisions

Cited by in corpus (6)

• Exploring Deep Neural Networks via Layer-Peeled Model: Minority Collapse in Imbalanced Training
• A Geometric Analysis of Neural Collapse with Unconstrained Features
• Image Clustering using an Augmented Generative Adversarial Network and Information Maximization
• Closed-Loop Data Transcription to an LDR via Minimaxing Rate Reduction
• Toward Better Generalization Bounds with Locally Elastic Stability
• Non-parametric Semi-Supervised Learning in Many-body Hilbert Space with Rescaled Logarithmic Fidelity