arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.00951cs.CVcs.AIcs.LG
收藏

与仿制品的反事实生成

Counterfactual Generation with Knockoffs

Friedrich-Schiller University(弗里德里希-席勒大学);Institute for Data Science of the German Aerospace Center(德国航空航天中心数据科学研究所)

Oana-Iuliana Popescu, Maha Shadaydeh, Joachim Denzler

摘要:

人类对深度神经网络决策的可解释性至关重要,特别是在直接影响人类生活的领域。已经训练好的神经网络的反事实解释可以通过扰动输入特征并根据扰动后分类器结果的变化来归属重要性来生成。扰动可以通过使用启发式或生成填充方法替换特征来完成。填充函数的选择显著影响伪影的数量,即,假阳性归因。探试法由于扰动后的图像远离原始数据分布而导致假阳性伪影。生成式填充方法通过生成尊重原始数据分布的填充值来减少伪影。然而,由于填充值与原始数据的高度相关性,当前的生成填充方法也可能增加假阴性。在本文中,我们建议通过使用基于遗传的Knockoffs框架生成填充来缓解这一问题,该框架由Barber和Candès在2015年开发,作为具有可控错误发现率的变量选择工具。敲除是尽可能与原始数据去相关的统计空变量,可以与原始数据交换而不改变底层数据分布。不同的填充方法的比较表明,填充与仿制品可以揭示更多的因果意义上的解释,同时仍然保持紧凑的解释。

英文摘要:

Human interpretability of deep neural networks' decisions is crucial, especially in domains where these directly affect human lives. Counterfactual explanations of already trained neural networks can be generated by perturbing input features and attributing importance according to the change in the classifier's outcome after perturbation. Perturbation can be done by replacing features using heuristic or generative in-filling methods. The choice of in-filling function significantly impacts the number of artifacts, i.e., false-positive attributions. Heuristic methods result in false-positive artifacts because the image after the perturbation is far from the original data distribution. Generative in-filling methods reduce artifacts by producing in-filling values that respect the original data distribution. However, current generative in-filling methods may also increase false-negatives due to the high correlation of in-filling values with the original data. In this paper, we propose to alleviate this by generating in-fillings with the statistically-grounded Knockoffs framework, which was developed by Barber and Candès in 2015 as a tool for variable selection with controllable false discovery rate. Knockoffs are statistically null-variables as decorrelated as possible from the original data, which can be swapped with the originals without changing the underlying data distribution. A comparison of different in-filling methods indicates that in-filling with knockoffs can reveal explanations in a more causal sense while still maintaining the compactness of the explanations.

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。

↑