反事实边际化:评估对干扰变量鲁棒性的框架
Counterfactual Marginalisation: Framework for Evaluating Robustness to Nuisance Variables
浏览论文内容
中文总结 AI 辅助
本文提出反事实边际化测试时评估框架,通过干预干扰变量生成反事实图像并平均预测,以量化分类模型对人口统计或采集捷径的鲁棒性,并定义风险、校准等指标。
中文摘要 AI 辅助
机器学习模型在依赖人口统计或采集相关捷径时,仍能实现较强的测试性能。我们提出反事实(CF)边际化作为一种测试时评估程序,用于评估分类模型对此类变量的鲁棒性。给定一个CF图像生成器,我们对年龄或性别等干扰父变量进行干预,生成每个测试图像的CF版本,并在目标干预分布上平均预测。这产生了干预感知的预测,在保留患者特定潜在信息的同时边际化人口统计效应。我们使用这些预测来定义CF风险、校准、稳定性和最坏情况敏感性的指标。我们展示了该框架在定量鲁棒性评估中的实用性。
英文摘要
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.
发表机构
- University of Oxford(牛津大学)
机构由 AI 辅助整理,请以论文原文为准。