Unmasking Clever Hans predictors and assessing what machines really learn
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Lapuschkin, Sebastian | - |
dc.contributor.author | Waeldchen, Stephan | - |
dc.contributor.author | Binder, Alexander | - |
dc.contributor.author | Montavon, Gregoire | - |
dc.contributor.author | Samek, Wojciech | - |
dc.contributor.author | Mueller, Klaus-Robert | - |
dc.date.accessioned | 2021-09-01T17:21:55Z | - |
dc.date.available | 2021-09-01T17:21:55Z | - |
dc.date.created | 2021-06-19 | - |
dc.date.issued | 2019-03-11 | - |
dc.identifier.issn | 2041-1723 | - |
dc.identifier.uri | https://scholar.korea.ac.kr/handle/2021.sw.korea/66681 | - |
dc.description.abstract | Current learning machines have successfully solved hard application problems, reaching high accuracy and displaying seemingly intelligent behavior. Here we apply recent techniques for explaining decisions of state-of-the-art learning machines and analyze various tasks from computer vision and arcade games. This showcases a spectrum of problem-solving behaviors ranging from naive and short-sighted, to well-informed and strategic. We observe that standard performance evaluation metrics can be oblivious to distinguishing these diverse problem solving behaviors. Furthermore, we propose our semi-automated Spectral Relevance Analysis that provides a practically effective way of characterizing and validating the behavior of nonlinear learning machines. This helps to assess whether a learned model indeed delivers reliably for the problem that it was conceived for. Furthermore, our work intends to add a voice of caution to the ongoing excitement about machine intelligence and pledges to evaluate and judge some of these recent successes in a more nuanced manner. | - |
dc.language | English | - |
dc.language.iso | en | - |
dc.publisher | NATURE PUBLISHING GROUP | - |
dc.subject | DEEP NEURAL-NETWORKS | - |
dc.subject | ARTIFICIAL-INTELLIGENCE | - |
dc.subject | CLASSIFICATION | - |
dc.subject | GO | - |
dc.subject | GAME | - |
dc.title | Unmasking Clever Hans predictors and assessing what machines really learn | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Mueller, Klaus-Robert | - |
dc.identifier.doi | 10.1038/s41467-019-08987-4 | - |
dc.identifier.scopusid | 2-s2.0-85062765505 | - |
dc.identifier.wosid | 000460759300001 | - |
dc.identifier.bibliographicCitation | NATURE COMMUNICATIONS, v.10 | - |
dc.relation.isPartOf | NATURE COMMUNICATIONS | - |
dc.citation.title | NATURE COMMUNICATIONS | - |
dc.citation.volume | 10 | - |
dc.type.rims | ART | - |
dc.type.docType | Article | - |
dc.description.journalClass | 1 | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Science & Technology - Other Topics | - |
dc.relation.journalWebOfScienceCategory | Multidisciplinary Sciences | - |
dc.subject.keywordPlus | DEEP NEURAL-NETWORKS | - |
dc.subject.keywordPlus | ARTIFICIAL-INTELLIGENCE | - |
dc.subject.keywordPlus | CLASSIFICATION | - |
dc.subject.keywordPlus | GO | - |
dc.subject.keywordPlus | GAME | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
(02841) 서울특별시 성북구 안암로 14502-3290-1114
COPYRIGHT © 2021 Korea University. All Rights Reserved.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.