Abstract

Deep learning has triggered the current rise of artificial intelligence and is the workhorse of today’s machine intelligence. Numerous success stories have rapidly spread all over science, industry and society, but its limitations have only recently come into focus. In this Perspective we seek to distil how many of deep learning’s failures can be seen as different symptoms of the same underlying problem: shortcut learning. Shortcuts are decision rules that perform well on standard benchmarks but fail to transfer to more challenging testing conditions, such as real-world scenarios. Related issues are known in comparative psychology, education and linguistics, suggesting that shortcut learning may be a common characteristic of learning systems, biological and artificial alike. Based on these observations, we develop a set of recommendations for model interpretation and benchmarking, highlighting recent advances in machine learning to improve robustness and transferability from the lab to real-world applications. Deep learning has resulted in impressive achievements, but under what circumstances does it fail, and why? The authors propose that its failures are a consequence of shortcut learning, a common characteristic across biological and artificial systems in which strategies that appear to have solved a problem fail unexpectedly under different circumstances.

Keywords

Artificial intelligenceComputer scienceDeep learningTransferabilityBenchmarkingMachine learningArtificial neural networkDeep neural networksRobustness (evolution)Perspective (graphical)Management

Affiliated Institutions

Related Publications

Publication Info

Year
2020
Type
article
Volume
2
Issue
11
Pages
665-673
Citations
1495
Access
Closed

Social Impact

Social media, news, blog, policy document mentions

Citation Metrics

1495
OpenAlex
118
Influential

Cite This

Robert Geirhos, Jörn-Henrik Jacobsen, Claudio Michaelis et al. (2020). Shortcut learning in deep neural networks. Nature Machine Intelligence , 2 (11) , 665-673. https://doi.org/10.1038/s42256-020-00257-z

Identifiers

DOI
10.1038/s42256-020-00257-z
arXiv
2004.07780

Data Quality

Data completeness: 84%