# Beyond traditional assumptions in fair machine learning

arXiv: https://arxiv.org/abs/2101.12476 · HTML: https://mlsanity.com/paper/2101.12476/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 212 references that are in the map.
- First posted on arXiv 29 Jan 2021.

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

- [Certifying and removing disparate impact](https://mlsanity.com/paper/1412.3756/) (2014)
- [Algorithmic decision making and the cost of fairness](https://mlsanity.com/paper/1701.08230/) (2017)
- [Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment](https://mlsanity.com/paper/1610.08452/) (2016)
- [Fairness in Criminal Justice Risk Assessments: The State of the Art](https://mlsanity.com/paper/1703.09207/) (2017)
- [Improving fairness in machine learning systems: What do industry practitioners need?](https://mlsanity.com/paper/1812.05239/) (2018)
- [Strategic Classification](https://mlsanity.com/paper/1506.06980/) (2015)
- [Fairness without Demographics through Adversarially Reweighted Learning](https://mlsanity.com/paper/2006.13114/) (2020)
- [A comparative study of fairness-enhancing interventions in machine learning](https://mlsanity.com/paper/1802.04422/) (2018)
- [Avoiding Discrimination through Causal Reasoning](https://mlsanity.com/paper/1706.02744/) (2017)
- [Foundations of Structural Causal Models with Cycles and Latent Variables](https://mlsanity.com/paper/1611.06221/) (2016)
- [The Social Cost of Strategic Classification](https://mlsanity.com/paper/1808.08460/) (2018)
- [Fairness Under Unawareness: Assessing Disparity When Protected Class Is Unobserved](https://mlsanity.com/paper/1811.11154/) (2018)
- [On Disentangled Representations Learned From Correlated Data](https://mlsanity.com/paper/2006.07886/) (2020)
- [Discrimination in Online Ad Delivery](https://mlsanity.com/paper/1301.6822/) (2013)
- [Operationalizing Individual Fairness with Pairwise Fair Representations](https://mlsanity.com/paper/1907.01439/) (2019)
- [Causal Reasoning for Algorithmic Fairness](https://mlsanity.com/paper/1805.05859/) (2018)
- [What's Sex Got To Do With Fair Machine Learning?](https://mlsanity.com/paper/2006.01770/) (2020)
- [From Fair Decision Making to Social Equality](https://mlsanity.com/paper/1812.02952/) (2018)
- [Generalization in anti-causal learning](https://mlsanity.com/paper/1812.00524/) (2018)
- [Policy Learning with Observational Data](https://mlsanity.com/paper/1702.02896/) (2017)
- [An Empirical Study on Learning Fairness Metrics for COMPAS Data with Human Supervision](https://mlsanity.com/paper/1910.10255/) (2019)
- [Removing Disparate Impact of Differentially Private Stochastic Gradient Descent on Model Accuracy](https://mlsanity.com/paper/2003.03699/) (2020)
- [Dimensions of Diversity in Human Perceptions of Algorithmic Fairness](https://mlsanity.com/paper/2005.00808/) (2020)
- [On the (im)possibility of fairness](https://mlsanity.com/paper/1609.07236/) (2016)
- [Causal Inference for Social Discrimination Reasoning](https://mlsanity.com/paper/1608.03735/) (2016)
- [Fair Inputs and Fair Outputs: The Incompatibility of Fairness in Privacy and Accuracy](https://mlsanity.com/paper/2005.09209/) (2020)
- [Exposing the Probabilistic Causal Structure of Discrimination](https://mlsanity.com/paper/1510.00552/) (2015)
- [Fair prediction with disparate impact: A study of bias in recidivism prediction instruments](https://mlsanity.com/paper/1703.00056/) (2017)
- [Exploration in two-stage recommender systems](https://mlsanity.com/paper/2009.08956/) (2020)
- [Algorithmic Decision Making in the Presence of Unmeasured Confounding](https://mlsanity.com/paper/1805.01868/) (2018)

## Built on by (most cited first, 2 of 2)

- [Talking About the Assumption in the Room](https://mlsanity.com/paper/2502.13268/) (2025)
- [Algorithmic Fairness Over Time: Advances & Prospects](https://mlsanity.com/paper/2610.15015/) (2026)

## Abstract (arXiv)

This thesis scrutinizes common assumptions underlying traditional machine learning approaches to fairness in consequential decision making. After challenging the validity of these assumptions in real-world applications, we propose ways to move forward when they are violated. First, we show that group fairness criteria purely based on statistical properties of observed data are fundamentally limited. Revisiting this limitation from a causal viewpoint we develop a more versatile conceptual framework, causal fairness criteria, and first algorithms to achieve them. We also provide tools to analyze how sensitive a believed-to-be causally fair algorithm is to misspecifications of the causal graph. Second, we overcome the assumption that sensitive data is readily available in practice. To this end we devise protocols based on secure multi-party computation to train, validate, and contest fair decision algorithms without requiring users to disclose their sensitive data or decision makers to disclose their models. Finally, we also accommodate the fact that outcome labels are often only observed when a certain decision has been made. We suggest a paradigm shift away from training predictive models towards directly learning decisions to relax the traditional assumption that labels can always be recorded. The main contribution of this thesis is the development of theoretically substantiated and practically feasible methods to move research on fair machine learning closer to real-world applications.