Revisiting the fragility of influence functions
Revisiting the fragility of influence functions
复制标题
重新审视影响力函数的脆弱性
DOI：
10.1016/j.neunet.2023.03.029
复制
发表时间：
2023
期刊：
影响因子：
7.8
通讯作者：
Rasool, Ghulam
中科院分区：
文献类型：
--
作者：
Epifano, Jacob R.;Ramachandran, Ravi P.;Masino, Aaron J.;Rasool, Ghulam
在过去的几年里，许多研究都试图解释深度学习模型的预测。然而，很少有人提出方法来验证这些解释的准确性或可靠性。影响函数是一种近似于“留一”训练对损失函数的影响的方法，近年来，影响函数已被证明是脆弱的。他们脆弱的原因尚不清楚。虽然以前的工作建议使用正则化来增加鲁棒性，但这并不适用于所有情况。在这项工作中，我们试图调查在之前的工作中进行的实验，以了解影响函数脆弱性的潜在机制。首先，我们在满足影响函数的凸性假设的条件下，使用文献中的程序验证影响函数。然后，我们放宽这些假设，并通过使用更深入的模型和更复杂的数据集来研究非凸性的影响。在这里，我们分析了用于验证影响函数的关键指标和程序。我们的结果表明，验证程序可能导致观察到的脆弱性。
In the last few years, many works have tried to explain the predictions of deep learning models. Few methods, however, have been proposed to verify the accuracy or faithfulness of these explanations. Recently, influence functions, which is a method that approximates the effect that leave-one-out training has on the loss function, has been shown to be fragile. The proposed reason for their fragility remains unclear. Although previous work suggests the use of regularization to increase robustness, this does not hold in all cases. In this work, we seek to investigate the experiments performed in the prior work in an effort to understand the underlying mechanisms of influence function fragility. First, we verify influence functions using procedures from the literature under conditions where the convexity assumptions of influence functions are met. Then, we relax these assumptions and study the effects of non-convexity by using deeper models and more complex datasets. Here, we analyze the key metrics and procedures that are used to validate influence functions. Our results indicate that the validation procedures may cause the observed fragility.