Invariant Learning via Probability of Sufficient and Necessary Causes

TL;DR

This paper introduces a causality-inspired approach for out-of-distribution generalization by leveraging the probability of sufficiency and necessity (PNS) of causes, improving model robustness across different domains.

Contribution

It proposes a novel PNS-based risk framework to learn invariant features considering both sufficiency and necessity, with theoretical guarantees and empirical validation.

Findings

PNS-based method improves OOD generalization on benchmarks.

Theoretical analysis confirms the generalizability of the PNS risk.

Experiments demonstrate effectiveness on synthetic and real-world data.

Abstract

Out-of-distribution (OOD) generalization is indispensable for learning models in the wild, where testing distribution typically unknown and different from the training. Recent methods derived from causality have shown great potential in achieving OOD generalization. However, existing methods mainly focus on the invariance property of causes, while largely overlooking the property of \textit{sufficiency} and \textit{necessity} conditions. Namely, a necessary but insufficient cause (feature) is invariant to distribution shift, yet it may not have required accuracy. By contrast, a sufficient yet unnecessary cause (feature) tends to fit specific data well but may have a risk of adapting to a new domain. To capture the information of sufficient and necessary causes, we employ a classical concept, the probability of sufficiency and necessary causes (PNS), which indicates the probability of whether one is the necessary and sufficient cause. To associate PNS with OOD generalization, we propose PNS risk and formulate an algorithm to learn representation with a high PNS value. We theoretically analyze and prove the generalizability of the PNS risk. Experiments on both synthetic and real-world benchmarks demonstrate the effectiveness of the proposed method. The details of the implementation can be found at the GitHub repository: https://github.com/ymy4323460/CaSN.