Learning how to explain neural networks: Patternnet and Patternattribution

Pieter Jan Kindermans, Kristof T. Schütt, Maximilian Alber, Klaus Robert Müller, Dumitru Erhan, Been Kim, Sven Dähne

Research output: Contribution to conferencePaper

17 Citations (Scopus)

Abstract

DeConvNet, Guided BackProp, LRP, were invented to better understand deep neural networks. We show that these methods do not produce the theoretically correct explanation for a linear model. Yet they are used on multi-layer networks with millions of parameters. This is a cause for concern since linear models are simple neural networks. We argue that explanation methods for neural nets should work reliably in the limit of simplicity, the linear models. Based on our analysis of linear models we propose a generalization that yields two explanation techniques (PatternNet and PatternAttribution) that are theoretically sound for linear models and produce improved explanations for deep networks.

Original languageEnglish
Publication statusPublished - 2018 Jan 1
Event6th International Conference on Learning Representations, ICLR 2018 - Vancouver, Canada
Duration: 2018 Apr 302018 May 3

Conference

Conference6th International Conference on Learning Representations, ICLR 2018
CountryCanada
CityVancouver
Period18/4/3018/5/3

    Fingerprint

ASJC Scopus subject areas

  • Language and Linguistics
  • Education
  • Computer Science Applications
  • Linguistics and Language

Cite this

Kindermans, P. J., Schütt, K. T., Alber, M., Müller, K. R., Erhan, D., Kim, B., & Dähne, S. (2018). Learning how to explain neural networks: Patternnet and Patternattribution. Paper presented at 6th International Conference on Learning Representations, ICLR 2018, Vancouver, Canada.