从标签偏差反推影响因素，用贝叶斯+方差分解更直观可靠。
A Bayesian explanation of machine learning models based on modes and functional ANOVA
- 基于贝叶斯框架逆向求解标签偏差的真正特征
- 通过ANOVA分解距离排序关键特征，效果优于平均值方法
- 计算成本与维度无关，适合高维数据解释
大多数可解释人工智能（XAI）方法聚焦于解释给定特征下的预测结果，而本文解决的是逆向解释问题：给定标签的偏离值，寻找导致该偏离的原因。我们采用贝叶斯框架，在观测标签值条件下恢复‘真实’特征。通过分析标签值偏离众数时的函数ANOVA分解中的‘距离’，高效识别并排序关键影响特征。实验表明，新方法比基于均值的方法（如SHAP值）更具人类直觉且更稳健。求解贝叶斯逆问题的额外开销与维度无关，具有良好的扩展性。
原文摘要 · Abstract (English)
Most methods in explainable AI (XAI) focus on providing reasons for the prediction of a given set of features. However, we solve an inverse explanation problem, i.e., given the deviation of a label, find the reasons of this deviation. We use a Bayesian framework to recover the ``true'' features, conditioned on the observed label value. We efficiently explain the deviation of a label value from the mode, by identifying and ranking the influential features using the ``distances'' in the ANOVA functional decomposition. We show that the new method is more human-intuitive and robust than methods based on mean values, e.g., SHapley Additive exPlanations (SHAP values). The extra costs of solving a Bayesian inverse problem are dimension-independent.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。