arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.07869cs.LGcs.CVcs.HCstat.ML
收藏

可解释性需要互动

Explainability Requires Interactivity

Hasso Plattner Institute for Digital Engineering, University of Potsdam(波茨坦大学哈索·普拉特纳数字工程研究所);TU Kaiserslautern(凯泽斯劳滕工业大学);Hasso Plattner Institute for Digital Health, Icahn School of Medicine at Mount Sinai(西奈山伊坎医学院哈索·普拉特纳数字健康研究所)

Matthias Kirchler, Martin Graf, Marius Kloft, Christoph Lippert

摘要:

在解释深层神经网络的决策时,简单的故事很诱人,但也很危险。特别是在计算机视觉中,最流行的解释方法给用户一种错误的理解感,并提供一个过于简单的图片。我们引入了一个交互式框架来理解现代视觉模型的高度复杂的决策边界。它允许用户彻底检查,探测和测试网络的决策。在一系列的案例研究中,我们比较了我们的交互式方法与静态解释方法的力量,展示了这些方法如何导致用户误入歧途,并可能带来严重的后果。

英文摘要:

When explaining the decisions of deep neural networks, simple stories are tempting but dangerous. Especially in computer vision, the most popular explanation approaches give a false sense of comprehension to its users and provide an overly simplistic picture. We introduce an interactive framework to understand the highly complex decision boundaries of modern vision models. It allows the user to exhaustively inspect, probe, and test a network's decisions. Across a range of case studies, we compare the power of our interactive approach to static explanation methods, showing how these can lead a user astray, with potentially severe consequences.

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。

↑