IDEAS home Printed from https://ideas.repec.org/a/nat/natcom/v10y2019i1d10.1038_s41467-019-08987-4.html
   My bibliography  Save this article

Unmasking Clever Hans predictors and assessing what machines really learn

Author

Listed:
  • 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
    Korea University
    Max Planck Institut für Informatik)

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.

Suggested Citation

  • Sebastian Lapuschkin & Stephan Wäldchen & Alexander Binder & Grégoire Montavon & Wojciech Samek & Klaus-Robert Müller, 2019. "Unmasking Clever Hans predictors and assessing what machines really learn," Nature Communications, Nature, vol. 10(1), pages 1-8, December.
  • Handle: RePEc:nat:natcom:v:10:y:2019:i:1:d:10.1038_s41467-019-08987-4
    DOI: 10.1038/s41467-019-08987-4
    as

    Download full text from publisher

    File URL: https://www.nature.com/articles/s41467-019-08987-4
    File Function: Abstract
    Download Restriction: no

    File URL: https://libkey.io/10.1038/s41467-019-08987-4?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Young Jae Kim & Jung-Im Na & Seung Seog Han & Chong Hyun Won & Mi Woo Lee & Jung-Won Shin & Chang-Hun Huh & Sung Eun Chang, 2022. "Augmenting the accuracy of trainee doctors in diagnosing skin lesions suspected of skin neoplasms in a real-world setting: A prospective controlled before-and-after study," PLOS ONE, Public Library of Science, vol. 17(1), pages 1-11, January.
    2. Jerome Friedman & Trevor Hastie & Robert Tibshirani, 2020. "Discussion of “Prediction, Estimation, and Attribution” by Bradley Efron," International Statistical Review, International Statistical Institute, vol. 88(S1), pages 73-74, December.
    3. Minji Lee & Leandro R. D. Sanz & Alice Barra & Audrey Wolff & Jaakko O. Nieminen & Melanie Boly & Mario Rosanova & Silvia Casarotto & Olivier Bodart & Jitka Annen & Aurore Thibaut & Rajanikant Panda &, 2022. "Quantifying arousal and awareness in altered states of consciousness using interpretable deep learning," Nature Communications, Nature, vol. 13(1), pages 1-14, December.
    4. Wang, Fujin & Zhao, Zhibin & Zhai, Zhi & Shang, Zuogang & Yan, Ruqiang & Chen, Xuefeng, 2023. "Explainability-driven model improvement for SOH estimation of lithium-ion battery," Reliability Engineering and System Safety, Elsevier, vol. 232(C).
    5. Minyoung Lee & Joohyoung Jeon & Hongchul Lee, 2022. "Explainable AI for domain experts: a post Hoc analysis of deep learning for defect classification of TFT–LCD panels," Journal of Intelligent Manufacturing, Springer, vol. 33(6), pages 1747-1759, August.
    6. Tobias Thomas & Dominik Straub & Fabian Tatai & Megan Shene & Tümer Tosik & Kristian Kersting & Constantin A. Rothkopf, 2024. "Modelling dataset bias in machine-learned theories of economic decision-making," Nature Human Behaviour, Nature, vol. 8(4), pages 679-691, April.
    7. Christoph March, 2019. "The Behavioral Economics of Artificial Intelligence: Lessons from Experiments with Computer Players," CESifo Working Paper Series 7926, CESifo.
    8. Krzysztof Fiok & Farzad V Farahani & Waldemar Karwowski & Tareq Ahram, 2022. "Explainable artificial intelligence for education and training," The Journal of Defense Modeling and Simulation, , vol. 19(2), pages 133-144, April.
    9. Xun Li & Dongsheng Chen & Weipan Xu & Haohui Chen & Junjun Li & Fan Mo, 2023. "Explainable dimensionality reduction (XDR) to unbox AI ‘black box’ models: A study of AI perspectives on the ethnic styles of village dwellings," Palgrave Communications, Palgrave Macmillan, vol. 10(1), pages 1-13, December.
    10. Verhagen, Mark D., 2021. "Identifying and Improving Functional Form Complexity: A Machine Learning Framework," SocArXiv bka76, Center for Open Science.
    11. Shane Fox & James McDermott & Edelle Doherty & Ronan Cooney & Eoghan Clifford, 2022. "Application of Neural Networks and Regression Modelling to Enable Environmental Regulatory Compliance and Energy Optimisation in a Sequencing Batch Reactor," Sustainability, MDPI, vol. 14(7), pages 1-28, March.
    12. March, Christoph, 2021. "Strategic interactions between humans and artificial intelligence: Lessons from experiments with computer players," Journal of Economic Psychology, Elsevier, vol. 87(C).
    13. Oliver T. Unke & Stefan Chmiela & Michael Gastegger & Kristof T. Schütt & Huziel E. Sauceda & Klaus-Robert Müller, 2021. "SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects," Nature Communications, Nature, vol. 12(1), pages 1-14, December.
    14. Diderich, Claude, 2023. "The Truth Behind Artificial Intelligence: Illustrated by Designing an Investment Advice Solution," Journal of Financial Transformation, Capco Institute, vol. 58, pages 116-125.
    15. Van Den Hauwe, Ludwig, 2023. "Why Machines Will Not Replace Entrepreneurs. On the Inevitable Limitations of Artificial Intelligence in Economic Life," MPRA Paper 118307, University Library of Munich, Germany.
    16. Martin Obschonka & David B. Audretsch, 2020. "Artificial intelligence and big data in entrepreneurship: a new era has begun," Small Business Economics, Springer, vol. 55(3), pages 529-539, October.

    More about this item

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:nat:natcom:v:10:y:2019:i:1:d:10.1038_s41467-019-08987-4. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    We have no bibliographic references for this item. You can help adding them by using this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.nature.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.