# On the Foundations of Shortcut Learning

arXiv: https://arxiv.org/abs/2310.16228 · HTML: https://mlsanity.com/paper/2310.16228/

## Key facts

- Cited by 65 papers in the map (data through 14 Oct 2026).
- Builds on 46 of its 58 references that are in the map.
- First posted on arXiv 24 Oct 2023.

## Builds on (most cited first, 30 of 46)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Neural Tangent Kernel: Convergence and Generalization in Neural Networks](https://mlsanity.com/paper/1806.07572/) (2018)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [On the Spectral Bias of Neural Networks](https://mlsanity.com/paper/1806.08734/) (2018)
- [SmoothGrad: removing noise by adding noise](https://mlsanity.com/paper/1706.03825/) (2017)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [RISE: Randomized Input Sampling for Explanation of Black-box Models](https://mlsanity.com/paper/1806.07421/) (2018)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [Recognition in Terra Incognita](https://mlsanity.com/paper/1807.04975/) (2018)
- [Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints](https://mlsanity.com/paper/1707.09457/) (2017)
- [Implicit Regularization in Deep Matrix Factorization](https://mlsanity.com/paper/1905.13655/) (2019)
- [Last Layer Re-Training is Sufficient for Robustness to Spurious Correlations](https://mlsanity.com/paper/2204.02937/) (2022)
- [Unmasking Clever Hans Predictors and Assessing What Machines Really Learn](https://mlsanity.com/paper/1902.10178/) (2019)
- [The Pitfalls of Simplicity Bias in Neural Networks](https://mlsanity.com/paper/2006.07710/) (2020)
- [Noise or Signal: The Role of Image Backgrounds in Object Recognition](https://mlsanity.com/paper/2006.09994/) (2020)
- [In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning](https://mlsanity.com/paper/1412.6614/) (2014)
- [An Investigation of Why Overparameterization Exacerbates Spurious Correlations](https://mlsanity.com/paper/2005.04345/) (2020)
- [A Fourier Perspective on Model Robustness in Computer Vision](https://mlsanity.com/paper/1906.08988/) (2019)
- [Gradient Starvation: A Learning Proclivity in Neural Networks](https://mlsanity.com/paper/2011.09468/) (2020)
- [Generalisation in humans and deep neural networks](https://mlsanity.com/paper/1808.08750/) (2018)
- [Deep learning generalizes because the parameter-function map is biased towards simple functions](https://mlsanity.com/paper/1805.08522/) (2018)
- [Understanding the Failure Modes of Out-of-Distribution Generalization](https://mlsanity.com/paper/2010.15775/) (2020)
- [What shapes feature representations? Exploring datasets, architectures, and training](https://mlsanity.com/paper/2006.12433/) (2020)
- [Can contrastive learning avoid shortcut solutions?](https://mlsanity.com/paper/2106.11230/) (2021)
- [Salient ImageNet: How to discover spurious features in Deep Learning?](https://mlsanity.com/paper/2110.04301/) (2021)
- [A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimation](https://mlsanity.com/paper/2306.07304/) (2023)

## Built on by (most cited first, 30 of 65)

- [SimBa: Simplicity Bias for Scaling Up Parameters in Deep Reinforcement Learning](https://mlsanity.com/paper/2410.09754/) (2024)
- [Neural Redshift: Random Networks are not Random Functions](https://mlsanity.com/paper/2403.02241/) (2024)
- [The Clever Hans Mirage: A Comprehensive Survey on Spurious Correlations in Machine Learning](https://mlsanity.com/paper/2402.12715/) (2024)
- [Benchmark Designers Should "Train on the Test Set" to Expose Exploitable Non-Visual Shortcuts](https://mlsanity.com/paper/2511.04655/) (2025)
- [All Patches Matter, More Patches Better: Enhance AI-Generated Image Detection via Panoptic Patch Learning](https://mlsanity.com/paper/2504.01396/) (2025)
- [Learned feature representations are biased by complexity, learning order, position, and more](https://mlsanity.com/paper/2405.05847/) (2024)
- [Dimensions underlying the representational alignment of deep neural networks with humans](https://mlsanity.com/paper/2406.19087/) (2024)
- [When does compositional structure yield compositional generalization? A kernel theory](https://mlsanity.com/paper/2405.16391/) (2024)
- [Decomposing and Editing Predictions by Modeling Model Computation](https://mlsanity.com/paper/2404.11534/) (2024)
- [Understanding Visual Feature Reliance through the Lens of Complexity](https://mlsanity.com/paper/2407.06076/) (2024)
- [Representation Learning and Identity Adversarial Training for Facial Behavior Understanding](https://mlsanity.com/paper/2407.11243/) (2024)
- [Spurious Correlations in High Dimensional Regression: The Roles of Regularization, Simplicity Bias and Over-Parameterization](https://mlsanity.com/paper/2502.01347/) (2025)
- [Bias in Motion: Theoretical Insights into the Dynamics of Bias in SGD Training](https://mlsanity.com/paper/2405.18296/) (2024)
- [Don't throw the baby out with the bathwater: How and why deep learning for ARC](https://mlsanity.com/paper/2506.14276/) (2025)
- [Towards Desiderata-Driven Design of Visual Counterfactual Explainers](https://mlsanity.com/paper/2506.14698/) (2025)
- [Mitigating Clever Hans Strategies in Image Classifiers through Generating Counterexamples](https://mlsanity.com/paper/2510.17524/) (2025)
- [Group-robust Sample Reweighting for Subpopulation Shifts via Influence Functions](https://mlsanity.com/paper/2503.07315/) (2025)
- [Representation biases: will we achieve complete understanding by analyzing representations?](https://mlsanity.com/paper/2507.22216/) (2025)
- [Rectifying Shortcut Behaviors in Preference-based Reward Learning](https://mlsanity.com/paper/2510.19050/) (2025)
- [Focus On This, Not That! Steering LLMs with Adaptive Feature Specification](https://mlsanity.com/paper/2410.22944/) (2024)
- [Physics-informed Temporal Alignment for Auto-regressive PDE Foundation Models](https://mlsanity.com/paper/2505.10930/) (2025)
- [Distinct Computations Emerge From Compositional Curricula in In-Context Learning](https://mlsanity.com/paper/2506.13253/) (2025)
- [Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training](https://mlsanity.com/paper/2605.11134/) (2026)
- [Rethinking and Red-Teaming Protective Perturbation in Personalized Diffusion Models](https://mlsanity.com/paper/2406.18944/) (2024)
- [Shortcut Learning Susceptibility in Vision Classifiers](https://mlsanity.com/paper/2502.09150/) (2025)
- [Preventing Shortcut Learning in Medical Image Analysis through Intermediate Layer Knowledge Distillation from Specialist Teachers](https://mlsanity.com/paper/2511.17421/) (2025)
- [From Invariant Representations to Invariant Data: Provable Robustness to Spurious Correlations via Noisy Counterfactual Matching](https://mlsanity.com/paper/2505.24843/) (2025)
- [Visual Anagrams Reveal Hidden Differences in Holistic Shape Processing Across Vision Models](https://mlsanity.com/paper/2507.00493/) (2025)
- [Beyond the Rosetta Stone: Unification Forces in Generalization Dynamics](https://mlsanity.com/paper/2508.11017/) (2025)
- [Fidelity-Aware Data Composition for Robust Robot Generalization](https://mlsanity.com/paper/2509.24797/) (2025)

## Abstract (arXiv)

Deep-learning models can extract a rich assortment of features from data. Which features a model uses depends not only on predictivity -- how reliably a feature indicates training-set labels -- but also on availability -- how easily the feature can be extracted from inputs. The literature on shortcut learning has noted examples in which models privilege one feature over another, for example texture over shape and image backgrounds over foreground objects. Here, we test hypotheses about which input properties are more available to a model, and systematically study how predictivity and availability interact to shape models' feature use. We construct a minimal, explicit generative framework for synthesizing classification datasets with two latent features that vary in predictivity and in factors we hypothesize to relate to availability, and we quantify a model's shortcut bias -- its over-reliance on the shortcut (more available, less predictive) feature at the expense of the core (less available, more predictive) feature. We find that linear models are relatively unbiased, but introducing a single hidden layer with ReLU or Tanh units yields a bias. Our empirical findings are consistent with a theoretical account based on Neural Tangent Kernels. Finally, we study how models used in practice trade off predictivity and availability in naturalistic datasets, discovering availability manipulations which increase models' degree of shortcut bias. Taken together, these findings suggest that the propensity to learn shortcut features is a fundamental characteristic of deep nonlinear architectures warranting systematic study given its role in shaping how models solve tasks.