English

Scalable Infomin Learning

Machine Learning 2023-02-22 v1 Artificial Intelligence Information Theory math.IT Machine Learning

Abstract

The task of infomin learning aims to learn a representation with high utility while being uninformative about a specified target, with the latter achieved by minimising the mutual information between the representation and the target. It has broad applications, ranging from training fair prediction models against protected attributes, to unsupervised learning with disentangled representations. Recent works on infomin learning mainly use adversarial training, which involves training a neural network to estimate mutual information or its proxy and thus is slow and difficult to optimise. Drawing on recent advances in slicing techniques, we propose a new infomin learning approach, which uses a novel proxy metric to mutual information. We further derive an accurate and analytically computable approximation to this proxy metric, thereby removing the need of constructing neural network-based mutual information estimators. Experiments on algorithmic fairness, disentangled representation learning and domain adaptation verify that our method can effectively remove unwanted information with limited time budget.

Keywords

Cite

@article{arxiv.2302.10701,
  title  = {Scalable Infomin Learning},
  author = {Yanzhi Chen and Weihao Sun and Yingzhen Li and Adrian Weller},
  journal= {arXiv preprint arXiv:2302.10701},
  year   = {2023}
}

Comments

10 pages, accepted to NeurIPS 2022, slightly improved version

R2 v1 2026-06-28T08:45:37.423Z