arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2001.05207cs.LGstat.ML
收藏

可解释性的形式化方法

A Formal Approach to Explainability

Facebook AI Research(脸书人工智能研究院);Tel Aviv University(特拉维夫大学);Technion(以色列理工学院)

Lior Wolf, Tomer Galanti, Tamir Hazan

摘要:

我们把解释看作是输入样本和模型输出的混合,并提供了一些定义,这些定义捕捉了生成这些解释的函数的各种所需属性。我们研究了这些属性之间的联系,以及学习模型的解释生成函数和中间表示之间的联系,并且能够证明,例如,如果给定层的激活与解释一致,那么所有其他后续层也是如此。此外,我们还研究了解释的交集和并集,作为构建新解释的一种方式。

英文摘要:

We regard explanations as a blending of the input sample and the model's output and offer a few definitions that capture various desired properties of the function that generates these explanations. We study the links between these properties and between explanation-generating functions and intermediate representations of learned models and are able to show, for example, that if the activations of a given layer are consistent with an explanation, then so do all other subsequent layers. In addition, we study the intersection and union of explanations as a way to construct new 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。

↑