Explaining nonlinear classification decisions with deep Taylor decomposition

作者：

Highlights：

• A novel method to explain nonlinear classification decisions in terms of input variables is introduced.

• The method is based on Taylor expansions and decomposes the output of a deep neural network in terms of input variables.

• The resulting deep Taylor decomposition can be applied directly to existing neural networks without retraining.

• The method is tested on two large-scale neural networks for image classification: BVLC CaffeNet and GoogleNet.

摘要

Highlights•A novel method to explain nonlinear classification decisions in terms of input variables is introduced.•The method is based on Taylor expansions and decomposes the output of a deep neural network in terms of input variables.•The resulting deep Taylor decomposition can be applied directly to existing neural networks without retraining.•The method is tested on two large-scale neural networks for image classification: BVLC CaffeNet and GoogleNet.

论文关键词：Deep neural networks,Heatmapping,Taylor decomposition,Relevance propagation,Image recognition

论文评审过程：Received 11 May 2016, Revised 8 August 2016, Accepted 12 November 2016, Available online 30 November 2016, Version of Record 5 January 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.11.008