arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.12935cs.LGcs.HC
收藏

时间序列模型归因可视化作为解释

Time Series Model Attribution Visualizations as Explanations

University of Konstanz(康斯坦茨大学)

Udo Schlegel, Daniel A. Keim

摘要:

归因是单个样本上深度学习模型的一种常见局部解释技术,因为它们很容易提取并展示输入值的相关性。在许多情况下,热图会将样本的此类属性可视化,例如在图像上。然而,热图并不总是解释其他数据类型的某些模型决策的理想可视化。在这篇评论中,我们专注于时间序列的归因可视化。我们收集了归因热图可视化和一些替代方案,讨论了其优点和缺点,并对未来的归因机会和时间序列的解释给出了一个简短的立场。

英文摘要:

Attributions are a common local explanation technique for deep learning models on single samples as they are easily extractable and demonstrate the relevance of input values. In many cases, heatmaps visualize such attributions for samples, for instance, on images. However, heatmaps are not always the ideal visualization to explain certain model decisions for other data types. In this review, we focus on attribution visualizations for time series. We collect attribution heatmap visualizations and some alternatives, discuss the advantages as well as disadvantages and give a short position towards future opportunities for attributions and explanations for time series.

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。

↑