归因评分中的逻辑陷阱
Logic Traps in Evaluating Attribution Scores
摘要：
现代深度学习模型是出了名的不透明，这促使人们开发了解释深度模型预测效果的方法。这一目标通常通过归因方法来实现，该方法评估特征对模型预测的影响。作为一种解释方法，归因方法的评价标准是它在多大程度上准确地反映了模型的实际推理过程（忠实性）。同时，由于深度模型的推理过程是不可访问的，研究人员设计了各种评估方法来证明他们的论点。然而，这些评价方法中的一些关键逻辑陷阱在大多数作品中被忽视，导致不准确的评价和不公平的比较。本文系统地回顾了现有的归因得分评估方法，并总结了这些方法中存在的逻辑陷阱。我们进一步进行实验来证明每个逻辑陷阱的存在。通过理论和实验分析，希望引起对归因分数不准确评价的关注。此外，在本文中，我们建议停止关注在不可靠的评估系统下提高性能，并开始努力减少拟议逻辑陷阱的影响
英文摘要：
Modern deep learning models are notoriously opaque, which has motivated the development of methods for interpreting how deep models predict. This goal is usually approached with attribution method, which assesses the influence of features on model predictions. As an explanation method, the evaluation criteria of attribution methods is how accurately it re-reflects the actual reasoning process of the model (faithfulness). Meanwhile, since the reasoning process of deep models is inaccessible, researchers design various evaluation methods to demonstrate their arguments. However, some crucial logic traps in these evaluation methods are ignored in most works, causing inaccurate evaluation and unfair comparison. This paper systematically reviews existing methods for evaluating attribution scores and summarizes the logic traps in these methods. We further conduct experiments to demonstrate the existence of each logic trap. Through both the theoretical and experimental analysis, we hope to increase attention on the inaccurate evaluation of attribution scores. Moreover, with this paper, we suggest stopping focusing on improving performance under unreliable evaluation systems and starting efforts on reducing the impact of proposed logic traps
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
