# Invariant Learning with Annotation-free Environments

arXiv: https://arxiv.org/abs/2504.15686 · HTML: https://mlsanity.com/paper/2504.15686/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 15 references that are in the map.
- First posted on arXiv 22 Apr 2025.

## Builds on (most cited first, 9 of 9)

- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)](https://mlsanity.com/paper/1902.03368/) (2019)
- [The Pitfalls of Simplicity Bias in Neural Networks](https://mlsanity.com/paper/2006.07710/) (2020)
- [Environment Inference for Invariant Learning](https://mlsanity.com/paper/2010.07249/) (2020)
- [Robustness to Spurious Correlations via Human Annotations](https://mlsanity.com/paper/2007.06661/) (2020)
- [DecAug: Out-of-Distribution Generalization via Decomposed Feature Representation and Semantic Augmentation](https://mlsanity.com/paper/2012.09382/) (2020)
- [An Overview of Melanoma Detection in Dermoscopy Images Using Image Processing and Machine Learning](https://mlsanity.com/paper/1601.07843/) (2016)
- [Out of Spuriousity: Improving Robustness to Spurious Correlations without Group Annotations](https://mlsanity.com/paper/2407.14974/) (2024)

## Built on by (most cited first, 1 of 1)

- [Deceptive Risk Minimization: Out-of-Distribution Generalization by Deceiving Distribution Shift Detectors](https://mlsanity.com/paper/2509.12081/) (2025)

## Abstract (arXiv)

Invariant learning is a promising approach to improve domain generalization compared to Empirical Risk Minimization (ERM). However, most invariant learning methods rely on the assumption that training examples are pre-partitioned into different known environments. We instead infer environments without the need for additional annotations, motivated by observations of the properties within the representation space of a trained ERM model. We show the preliminary effectiveness of our approach on the ColoredMNIST benchmark, achieving performance comparable to methods requiring explicit environment labels and on par with an annotation-free method that poses strong restrictions on the ERM reference model.