Counterfactual Marginalisation: Framework for Evaluating Robustness to Nuisance Variables

2026-09-09

AI Summary

提出了一种基于反事实边际化的测试时间评估方法，通过生成反事实图像并平均预测来评估分类模型对干扰变量的鲁棒性。

Abstract

Machine learning models can achieve strong test performance while relying on demographic or acquisition-related shortcuts. We propose counterfactual (CF) marginalisation as a test-time evaluation procedure for assessing robustness of classification models to such variables. Given a CF image generator, we intervene on nuisance parent variables such as age or sex, generate CF versions of each test image, and average predictions over a target intervention distribution. This produces intervention-aware predictions that marginalise demographic effects while preserving patient-specific latent information. We use these predictions to define metrics for CF risk, calibration, stability and worst-case sensitivity. We demonstrate this framework's utility for quantitative robustness evaluation.

Problem

• robustness
• nuisance variables
• counterfactual
• classification models
• demographic

Innovation

• counterfactual marginalisation
• nuisance variables
• robustness evaluation
• intervention-aware predictions