Refining the Information Bottleneck via Adversarial Information Separation

TL;DR

This paper introduces AdverISF, a novel adversarial framework that effectively separates task-relevant features from noise without supervision, improving generalization in data-scarce and real-world material design tasks.

Contribution

The paper presents a self-supervised adversarial method with a multi-layer architecture for unsupervised feature separation, advancing beyond existing techniques.

Findings

Outperforms state-of-the-art methods in data-scarce scenarios

Achieves superior generalization in material design tasks

Enables finer-grained feature extraction through hierarchical separation

Abstract

Generalizing from limited data is particularly critical for models in domains such as material science, where task-relevant features in experimental datasets are often heavily confounded by measurement noise and experimental artifacts. Standard regularization techniques fail to precisely separate meaningful features from noise, while existing adversarial adaptation methods are limited by their reliance on explicit separation labels. To address this challenge, we propose the Adversarial Information Separation Framework (AdverISF), which isolates task-relevant features from noise without requiring explicit supervision. AdverISF introduces a self-supervised adversarial mechanism to enforce statistical independence between task-relevant features and noise representations. It further employs a multi-layer separation architecture that progressively recycles noise information across feature hierarchies to recover features inadvertently discarded as noise, thereby enabling finer-grained feature extraction. Extensive experiments demonstrate that AdverISF outperforms state-of-the-art methods in data-scarce scenarios. In addition, evaluations on real-world material design tasks show that it achieves superior generalization performance.