用对抗扰动替代像素修改，提升卷积网络解释图评估可靠性
Reliable Evaluation of Attribution Maps in CNNs: A Perturbation-Based Approach
- 以对抗扰动替代传统插入/删除操作，增强评估鲁棒性
- 在15个数据集-模型组合上验证，排名一致性显著提升
- 首次全面评测16种解释方法，SmoothGrad表现最优
本文提出一种评估卷积神经网络解释图的新方法。针对现有插入/删除指标易受分布偏移影响的问题，采用对抗扰动替代像素修改，结合平滑性和单调性度量，有效修正分布偏移带来的偏差。我们进行了迄今最全面的定量与定性评估，引入基线解释图作为合理性检验，结果表明只有本方法通过所有检验。基于肯德尔τ相关系数，在15个数据集-架构组合中，本方法显著提升排名一致性。对16种解释方法的测试表明，SmoothGrad是当前表现最佳的解释图。研究为解释图的可靠评估提供了统一框架，并将公开代码以保障可复现性。
原文摘要 · Abstract (English)
In this paper, we present an approach for evaluating attribution maps, which play a central role in interpreting the predictions of convolutional neural networks (CNNs). We show that the widely used insertion/deletion metrics are susceptible to distribution shifts that affect the reliability of the ranking. Our method proposes to replace pixel modifications with adversarial perturbations, which provides a more robust evaluation framework. By using smoothness and monotonicity measures, we illustrate the effectiveness of our approach in correcting distribution shifts. In addition, we conduct the most comprehensive quantitative and qualitative assessment of attribution maps to date. Introducing baseline attribution maps as sanity checks, we find that our metric is the only contender to pass all checks. Using Kendall's $τ$ rank correlation coefficient, we show the increased consistency of our metric across 15 dataset-architecture combinations. Of the 16 attribution maps tested, our results clearly show SmoothGrad to be the best map currently available. This research makes an important contribution to the development of attribution maps by providing a reliable and consistent evaluation framework. To ensure reproducibility, we will provide the code along with our results.
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。