# A Primer on Causal and Statistical Dataset Biases for Fair and Robust Image Analysis

arXiv: https://arxiv.org/abs/2509.04295 · HTML: https://mlsanity.com/paper/2509.04295/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 27 references that are in the map.
- First posted on arXiv 4 Sep 2025.

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

- [Equality of Opportunity in Supervised Learning](https://mlsanity.com/paper/1610.02413/) (2016)
- [Understanding deep learning requires rethinking generalization](https://mlsanity.com/paper/1611.03530/) (2016)
- [Variational Autoencoders and Nonlinear ICA: A Unifying Framework](https://mlsanity.com/paper/1907.04809/) (2019)
- [Learning Not to Learn: Training Deep Neural Networks with Biased Data](https://mlsanity.com/paper/1812.10352/) (2018)
- [Reading Race: AI Recognises Patient's Racial Identity In Medical Images](https://mlsanity.com/paper/2107.10356/) (2021)
- [MEDFAIR: Benchmarking Fairness for Medical Imaging](https://mlsanity.com/paper/2210.01725/) (2022)
- [Causal Networks: Semantics and Expressiveness](https://mlsanity.com/paper/1304.2379/) (2013)
- [Identifiability of latent-variable and structural-equation models: from linear to nonlinear](https://mlsanity.com/paper/2302.02672/) (2023)
- [Leveling Down in Computer Vision: Pareto Inefficiencies in Fair Deep Classifiers](https://mlsanity.com/paper/2203.04913/) (2022)
- [Improving the Fairness of Chest X-ray Classifiers](https://mlsanity.com/paper/2203.12609/) (2022)
- [An Empirical Characterization of Fair Machine Learning For Clinical Risk Prediction](https://mlsanity.com/paper/2007.10306/) (2020)
- [The Unfairness of Fair Machine Learning: Levelling down and strict egalitarianism by default](https://mlsanity.com/paper/2302.02404/) (2023)
- [Potential sources of dataset bias complicate investigation of underdiagnosis by machine learning algorithms](https://mlsanity.com/paper/2201.07856/) (2022)
- [The Role of Subgroup Separability in Group-Fair Medical Image Classification](https://mlsanity.com/paper/2307.02791/) (2023)
- [Rethinking Fair Representation Learning for Performance-Sensitive Tasks](https://mlsanity.com/paper/2410.04120/) (2024)

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

- [Effect of Demographic Bias on Skin Lesion Classification](https://mlsanity.com/paper/2606.03214/) (2026)

## Abstract (arXiv)

Machine learning methods often fail when deployed in the real world. Worse still, they fail in high-stakes situations and across socially sensitive lines. These issues have a chilling effect on the adoption of machine learning methods in settings such as medical diagnosis, where they are arguably best-placed to provide benefits if safely deployed. In this primer, we introduce the causal and statistical structures which induce failure in machine learning methods for image analysis. We highlight two previously overlooked problems, which we call the textitno fair lunch problem and the textitsubgroup separability problem. We elucidate why today's fair representation learning methods fail to adequately solve them and propose potential paths forward for the field.