Unmasking Clever Hans predictors and assessing what machines really learn

Sebastian Lapuschkin, Stephan Wäldchen, Alexander Binder, Grégoire Montavon, Wojciech Samek () and Klaus-Robert Müller ()
Additional contact information
Sebastian Lapuschkin: Fraunhofer Heinrich Hertz Institute
Stephan Wäldchen: Technische Universität Berlin
Alexander Binder: Singapore University of Technology and Design
Grégoire Montavon: Technische Universität Berlin
Wojciech Samek: Fraunhofer Heinrich Hertz Institute
Klaus-Robert Müller: Technische Universität Berlin

Nature Communications, 2019, vol. 10, issue 1, 1-8

Abstract: 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.

Date: 2019
References: Add references at CitEc
Citations: View citations in EconPapers (17)

Downloads: (external link)
https://www.nature.com/articles/s41467-019-08987-4 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:10:y:2019:i:1:d:10.1038_s41467-019-08987-4

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-019-08987-4

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().