# Monitoring Shortcut Learning using Mutual Information

arXiv: https://arxiv.org/abs/2206.13034 · HTML: https://mlsanity.com/paper/2206.13034/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 27 references that are in the map.
- First posted on arXiv 27 Jun 2022.

## Builds on (most cited first, 19 of 19)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Neural Tangent Kernel: Convergence and Generalization in Neural Networks](https://mlsanity.com/paper/1806.07572/) (2018)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Deep Learning and the Information Bottleneck Principle](https://mlsanity.com/paper/1503.02406/) (2015)
- [Opening the Black Box of Deep Neural Networks via Information](https://mlsanity.com/paper/1703.00810/) (2017)
- [Causality matters in medical imaging](https://mlsanity.com/paper/1912.08142/) (2019)
- [Qualitatively characterizing neural network optimization problems](https://mlsanity.com/paper/1412.6544/) (2014)
- [A Fine-Grained Analysis on Distribution Shift](https://mlsanity.com/paper/2110.11328/) (2021)
- [Strike (with) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects](https://mlsanity.com/paper/1811.11553/) (2018)
- [i-RevNet: Deep Invertible Networks](https://mlsanity.com/paper/1802.07088/) (2018)
- [Dark Model Adaptation: Semantic Image Segmentation from Daytime to Nighttime](https://mlsanity.com/paper/1810.02575/) (2018)
- [Unsupervised domain adaptation for medical imaging segmentation with self-ensembling](https://mlsanity.com/paper/1811.06042/) (2018)
- [Evaluating Saliency Map Explanations for Convolutional Neural Networks: A User Study](https://mlsanity.com/paper/2002.00772/) (2020)
- [Entropy and mutual information in models of deep neural networks](https://mlsanity.com/paper/1805.09785/) (2018)
- [Deep Learning Predicts Hip Fracture using Confounding Patient and Healthcare Variables](https://mlsanity.com/paper/1811.03695/) (2018)
- [Are Visual Explanations Useful? A Case Study in Model-in-the-Loop Prediction](https://mlsanity.com/paper/2007.12248/) (2020)
- [Information in Infinite Ensembles of Infinitely-Wide Neural Networks](https://mlsanity.com/paper/1911.09189/) (2019)

## Built on by (most cited first, 6 of 6)

- [Auxiliary Losses for Learning Generalizable Concept-based Models](https://mlsanity.com/paper/2311.11108/) (2023)
- [Demonstrating and Reducing Shortcuts in Vision-Language Representation Learning](https://mlsanity.com/paper/2402.17510/) (2024)
- [Beyond Distribution Shift: Spurious Features Through the Lens of Training Dynamics](https://mlsanity.com/paper/2302.09344/) (2023)
- [Exploiting Supervised Poison Vulnerability to Strengthen Self-Supervised Defense](https://mlsanity.com/paper/2409.08509/) (2024)
- [Trust Me, I Know the Way: Predictive Uncertainty in the Presence of Shortcut Learning](https://mlsanity.com/paper/2502.09137/) (2025)
- [Diagnosing Shortcut-Induced Rigidity in Continual Learning: The Einstellung Rigidity Index (ERI)](https://mlsanity.com/paper/2510.00475/) (2025)

## Abstract (arXiv)

The failure of deep neural networks to generalize to out-of-distribution data is a well-known problem and raises concerns about the deployment of trained networks in safety-critical domains such as healthcare, finance and autonomous vehicles. We study a particular kind of distribution shift unicodex2013 shortcuts or spurious correlations in the training data. Shortcut learning is often only exposed when models are evaluated on real-world data that does not contain the same spurious correlations, posing a serious dilemma for AI practitioners to properly assess the effectiveness of a trained model for real-world applications. In this work, we propose to use the mutual information (MI) between the learned representation and the input as a metric to find where in training, the network latches onto shortcuts. Experiments demonstrate that MI can be used as a domain-agnostic metric for monitoring shortcut learning.