{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-how-to-explain-neural-networks","title":"Learning how to explain neural networks: PatternNet and PatternAttribution","arxiv_id":"1705.05598","date":"2017-05-16","proceeding":"ICLR 2018 1","authors":["Pieter-Jan Kindermans","Kristof T. Schütt","Maximilian Alber","Klaus-Robert Müller","Dumitru Erhan","Been Kim","Sven Dähne"],"abstract":"DeConvNet, Guided BackProp, LRP, were invented to better understand deep\nneural networks. We show that these methods do not produce the theoretically\ncorrect explanation for a linear model. Yet they are used on multi-layer\nnetworks with millions of parameters. This is a cause for concern since linear\nmodels are simple neural networks. We argue that explanation methods for neural\nnets should work reliably in the limit of simplicity, the linear models. Based\non our analysis of linear models we propose a generalization that yields two\nexplanation techniques (PatternNet and PatternAttribution) that are\ntheoretically sound for linear models and produce improved explanations for\ndeep networks.","url_abs":"http://arxiv.org/abs/1705.05598v2","url_pdf":"http://arxiv.org/pdf/1705.05598v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-how-to-explain-neural-networks","repo_url":"https://github.com/DFKI-NLP/language-attributions","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-how-to-explain-neural-networks","repo_url":"https://github.com/adrhill/explainableai.jl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-how-to-explain-neural-networks","repo_url":"https://github.com/dianna-ai/dianna","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"learning-how-to-explain-neural-networks","repo_url":"https://github.com/pikinder/nn-patterns","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.05598","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}