arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.09949cs.LGcs.CV
收藏

SpecXAI --深度学习模型的谱可解释性

SpecXAI -- Spectral interpretability of Deep Learning Models

Monolith AI LTD.(Monolith AI有限公司);Iowa State University(爱荷华州立大学)

Stefan Druc, Peter Wooldridge, Adarsh Krishnamurthy, Soumik Sarkar, Aditya Balu

摘要:

深度学习由于能够将大量数据转换为高性能模型,因此越来越多地被商业和工业采用。然而,这些模型通常被认为是黑匣子,尽管它们的性能很好,但可能会阻止它们的使用。在这种背景下,可解释的人工智能领域试图开发一些技术,以缓和模型的不可穿透性,并促进对其行为的理解。在这里,我们以一个框架的形式提出了我们对XAI方法的贡献,我们称之为SpecXAI,它是基于整个网络的光谱表征。我们展示了如何使用这个框架不仅可以理解网络,而且还可以将其操作成线性可解释的符号表示。

英文摘要:

Deep learning is becoming increasingly adopted in business and industry due to its ability to transform large quantities of data into high-performing models. These models, however, are generally regarded as black boxes, which, in spite of their performance, could prevent their use. In this context, the field of eXplainable AI attempts to develop techniques that temper the impenetrable nature of the models and promote a level of understanding of their behavior. Here we present our contribution to XAI methods in the form of a framework that we term SpecXAI, which is based on the spectral characterization of the entire network. We show how this framework can be used to not only understand the network but also manipulate it into a linear interpretable symbolic representation.

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。

↑