Information Subtraction: Learning Representations for Conditional Entropy

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Leong, Keng Hou, Xiu, Yuxuan, Kin, Wai, Chan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913634998812672
author Leong, Keng Hou
Xiu, Yuxuan
Kin, Wai
Chan
author_facet Leong, Keng Hou
Xiu, Yuxuan
Kin, Wai
Chan
contents The representations of conditional entropy and conditional mutual information are significant in explaining the unique effects among variables. While previous studies based on conditional contrastive sampling have effectively removed information regarding discrete sensitive variables, they have not yet extended their scope to continuous cases. This paper introduces Information Subtraction, a framework designed to generate representations that preserve desired information while eliminating the undesired. We implement a generative-based architecture that outputs these representations by simultaneously maximizing an information term and minimizing another. With its flexibility in disentangling information, we can iteratively apply Information Subtraction to represent arbitrary information components between continuous variables, thereby explaining the various relationships that exist between them. Our results highlight the representations' ability to provide semantic features of conditional entropy. By subtracting sensitive and domain-specific information, our framework demonstrates effective performance in fair learning and domain generalization. The code for this paper is available at https://github.com/jh-liang/Information-Subtraction
format Preprint
id arxiv_https___arxiv_org_abs_2501_02012
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Information Subtraction: Learning Representations for Conditional Entropy
Leong, Keng Hou
Xiu, Yuxuan
Kin, Wai
Chan
Machine Learning
The representations of conditional entropy and conditional mutual information are significant in explaining the unique effects among variables. While previous studies based on conditional contrastive sampling have effectively removed information regarding discrete sensitive variables, they have not yet extended their scope to continuous cases. This paper introduces Information Subtraction, a framework designed to generate representations that preserve desired information while eliminating the undesired. We implement a generative-based architecture that outputs these representations by simultaneously maximizing an information term and minimizing another. With its flexibility in disentangling information, we can iteratively apply Information Subtraction to represent arbitrary information components between continuous variables, thereby explaining the various relationships that exist between them. Our results highlight the representations' ability to provide semantic features of conditional entropy. By subtracting sensitive and domain-specific information, our framework demonstrates effective performance in fair learning and domain generalization. The code for this paper is available at https://github.com/jh-liang/Information-Subtraction
title Information Subtraction: Learning Representations for Conditional Entropy
topic Machine Learning
url https://arxiv.org/abs/2501.02012