Variational Information Maximization for Feature Selection

Abstract

Feature selection is one of the most fundamental problems in machine learning. An extensive body of work on information-theoretic feature selection exists which is based on maximizing mutual information between subsets of features and class labels. Practical methods are forced to rely on approximations due to the difficulty of estimating mutual information. We demonstrate that approximations made by existing methods are based on unrealistic assumptions. We formulate a more flexible and general class of assumptions based on variational distributions and use them to tractably generate lower bounds for mutual information. These bounds define a novel information-theoretic framework for feature selection, which we prove to be optimal under tree graphical models with proper choice of variational distributions. Our experiments demonstrate that the proposed method strongly outperforms existing information-theoretic feature selection approaches.

15 pages, 9 figures

References in corpus (2)

• Submodular meets Spectral: Greedy Algorithms for Subset Selection, Sparse Approximation and Dictionary Selection
• Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning

Cited by in corpus (8)

• Learning to Explain: An Information-Theoretic Perspective on Model Interpretation
• When is invariance useful in an Out-of-Distribution Generalization problem ?
• Generative causal explanations of black-box classifiers
• Discovery and Separation of Features for Invariant Representation Learning
• Feature Selection Based on Unique Relevant Information for Health Data
• Modeling Psychotherapy Dialogues with Kernelized Hashcode Representations: A Nonparametric Information-Theoretic Approach
• Kernelized Hashcode Representations for Relation Extraction
• Less is More: Feature Selection for Adversarial Robustness with Compressive Counter-Adversarial Attacks