arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.03759cs.AI
收藏

专注时间:使用时间序列归因方法的综合基准

Time to Focus: A Comprehensive Benchmark Using Time Series Attribution Methods

German Research Center for Artificial Intelligence GmbH (DFKI)(德国人工智能研究中心有限公司);Technical University Kaiserslautern (TUK)(凯泽斯劳滕工业大学)

Dominique Mercier, Jwalin Bhatt, Andreas Dengel, Sheraz Ahmed

摘要:

在过去的十年里,神经网络在工业和研究领域都产生了巨大的影响,因为它们能够从不精确或复杂的数据中提取有意义的特征,并且在几个领域实现了超越人类的性能。然而,由于缺乏透明度,这些网络的使用在安全关键领域受到阻碍。在安全关键领域,这是法律所必需的。最近,已经提出了几种方法来揭示这个黑匣子,通过提供这些模型的预测解释。本文的重点是时间序列分析和基准几个国家的最先进的归因方法计算卷积分类器的解释。所提出的实验涉及基于梯度和扰动的归因方法。详细的分析表明,扰动为基础的方法是优越的敏感性和闭塞的游戏。这些方法倾向于产生具有更高连续性的解释。首先,基于梯度的技术在运行时间和失真方面都很出色。此外,验证的方法的依赖于训练的模型,可行的应用领域,和个人的特点。研究结果强调,选择最合适的归因方法与所需的用例密切相关。无论是哪一类归因方法,还是哪一种单一方法,都没有在所有方面表现出出色的表现。

英文摘要:

In the last decade neural network have made huge impact both in industry and research due to their ability to extract meaningful features from imprecise or complex data, and by achieving super human performance in several domains. However, due to the lack of transparency the use of these networks is hampered in the areas with safety critical areas. In safety-critical areas, this is necessary by law. Recently several methods have been proposed to uncover this black box by providing interpreation of predictions made by these models. The paper focuses on time series analysis and benchmark several state-of-the-art attribution methods which compute explanations for convolutional classifiers. The presented experiments involve gradient-based and perturbation-based attribution methods. A detailed analysis shows that perturbation-based approaches are superior concerning the Sensitivity and occlusion game. These methods tend to produce explanations with higher continuity. Contrarily, the gradient-based techniques are superb in runtime and Infidelity. In addition, a validation the dependence of the methods on the trained model, feasible application domains, and individual characteristics is attached. The findings accentuate that choosing the best-suited attribution method is strongly correlated with the desired use case. Neither category of attribution methods nor a single approach has shown outstanding performance across all aspects.

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。

↑