# The Importance of Background Information for Out of Distribution Generalization

arXiv: https://arxiv.org/abs/2206.08794 · HTML: https://mlsanity.com/paper/2206.08794/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 22 references that are in the map.
- First posted on arXiv 17 Jun 2022.

## Builds on (most cited first, 15 of 15)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [In Search of Lost Domain Generalization](https://mlsanity.com/paper/2007.01434/) (2020)
- [MIMIC-CXR-JPG, a large publicly available database of labeled chest radiographs](https://mlsanity.com/paper/1901.07042/) (2019)
- [Confounding variables can degrade generalization performance of radiological deep learning models](https://mlsanity.com/paper/1807.00431/) (2018)
- [Unified Deep Supervised Domain Adaptation and Generalization](https://mlsanity.com/paper/1709.10190/) (2017)
- [Right for the Right Reasons: Training Differentiable Models by Constraining their Explanations](https://mlsanity.com/paper/1703.03717/) (2017)
- [Learning explanations that are hard to vary](https://mlsanity.com/paper/2009.00329/) (2020)
- [What shapes feature representations? Exploring datasets, architectures, and training](https://mlsanity.com/paper/2006.12433/) (2020)
- [Self-training Avoids Using Spurious Features Under Domain Shift](https://mlsanity.com/paper/2006.10032/) (2020)
- [GradMask: Reduce Overfitting by Regularizing Saliency](https://mlsanity.com/paper/1904.07478/) (2019)
- [Saliency is a Possible Red Herring When Diagnosing Poor Generalization](https://mlsanity.com/paper/1910.00199/) (2019)

## Abstract (arXiv)

Domain generalization in medical image classification is an important problem for trustworthy machine learning to be deployed in healthcare. We find that existing approaches for domain generalization which utilize ground-truth abnormality segmentations to control feature attributions have poor out-of-distribution (OOD) performance relative to the standard baseline of empirical risk minimization (ERM). We investigate what regions of an image are important for medical image classification and show that parts of the background, that which is not contained in the abnormality segmentation, provides helpful signal. We then develop a new task-specific mask which covers all relevant regions. Utilizing this new segmentation mask significantly improves the performance of the existing methods on the OOD test sets. To obtain better generalization results than ERM, we find it necessary to scale up the training data size in addition to the usage of these task-specific masks.