检验了三种解释方法在表格数据上的可信度，发现模型表现好不一定解释可靠。
Evaluating Local Explainability Metrics for Machine Learning Models on Tabular Data
- 对比LIME、SHAP和特征消融法在32个数据集上的解释效果
- 解释质量与模型预测性能无关，受数据复杂度和特征分布影响大
- 适合关注模型解释可信度的研究者或工程师参考
尽管解释技术被广泛用于理解人工智能行为，但生成的解释未必可靠。一个看似合理的解释可能并未捕捉模型内部推理机制，尤其是在处理复杂表格数据时。本文研究了局部解释技术在复杂表格分类任务中的可信度，评估了三类核心属性：对模型预测的忠实性、对输入变化的鲁棒性以及解释本身的复杂度。在32个数据集上对局部可解释模型无关解释（LIME）、核谢尔利加性解释（SHAP）和特征消融法进行了基准测试，涵盖不同类型的机器学习模型。分析了模型性能范围，识别出两类样本：共识正确（所有模型均预测正确）和共识错误（所有模型均预测错误）。结果表明，解释质量并不总是与模型预测性能相关，数据复杂度和特征分布才是影响解释质量和可靠性的主要因素。
原文摘要 · Abstract (English)
Despite the wide use of explainability techniques to attempt to understand the behavior of Artificial Intelligence (AI), the generated explanations may not always be reliable. An explanation can appear plausible to humans but fail to capture the internal reasoning of a model, particularly when dealing with complex tabular data. This paper studies the trustworthiness of local explainability techniques when applied to complex tabular classification tasks, considering evaluated metrics for three main properties: faithfulness to the model's predictions, robustness to input data variations, and complexity of the explanation itself. A benchmark was performed for Local Interpretable Model-Agnostic Explanations (LIME), Kernel SHapley Additive exPlanations (SHAP), and Feature Ablation techniques, across 32 datasets and different types of machine learning models. Model performance ranges were analyzed to identify two groups: consensus-correct, which are samples that all models predicted correctly, and consensus-wrong, samples that all models predicted incorrectly. The obtained results demonstrate that that the explanations are not always correlated with a model's predictive performance. Instead, dataset complexity and feature distributions seem to be the main factors affecting explanation quality and reliability.
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。