Machine learning (ML) systems are utilized in critical sectors such as healthcare, law enforcement, and finance, but often rely on historical data that contains demographic biases, leading to decisions that perpetuate or intensify existing inequalities. Causal and counterfactual fairness provide a transparent, human-in-the-loop framework to mitigate algorithmic discrimination, aligning closely with legal doctrines of direct and indirect discrimination. However, current causal fairness frameworks hold a key limitation in that they assume prior knowledge of the correct causal model, restricting their applicability in complex fairness scenarios where causal models are unknown or difficult to identify. To bridge this gap, we propose FairPFN, a tabular foundation model pre-trained on synthetic causal fairness data to identify and mitigate the causal effects of protected attributes in its predictions. FairPFN’s key contribution is that it requires no knowledge of the causal model and demonstrates strong performance across a diverse set of hand-crafted and real-world causal scenarios relative to robust baseline methods. FairPFN paves the way for a promising direction for future research, making causal fairness more accessible to a wider variety of complex fairness problems.

Causal effect estimation is critical to a range of scientific disciplines. Existing methods for this task either require interventional data, knowledge about the ground-truth causal graph, or rely on assumptions such as unconfoundedness, restricting their applicability in real-world settings. In the domain of tabular machine learning, Prior-data fitted networks (PFNs) have achieved state-of-the-art predictive performance, having been pre-trained on synthetic causal data to solve tabular prediction problems via in-context learning. To assess whether this can be transferred to the problem of causal effect estimation, we pre-train PFNs on synthetic data drawn from a wide variety of causal structures, including interventions, to predict interventional outcomes given observational data. Through extensive experiments in synthetic and semi-synthetic settings, we show that our approach allows for the accurate estimation of causal effects without knowledge of the underlying causal graph.

Fairness-aware Machine Learning (FairML) applications are often characterized by complex social objectives and legal requirements, frequently involving multiple, potentially conflicting notions of fairness. Despite the well-known Impossibility Theorem of Fairness and extensive theoretical research on the statistical and socio-technical trade-offs between fairness metrics, many FairML tools still optimize or constrain for a single fairness objective. However, this one-sided optimization can inadvertently lead to violations of other relevant notions of fairness. In this socio-technical and empirical study, we frame fairness as a Many-Objective (MaO) problem by treating fairness metrics as conflicting objectives in a multi-objective (MO) sense. We introduce ManyFairHPO, a human-in-the-loop, fairness-aware model selection framework that enables practitioners to effectively navigate complex and nuanced fairness objective landscapes. ManyFairHPO aids in the identification, evaluation, and balancing of fairness metric conflicts and their related social consequences, leading to more informed and socially responsible model-selection decisions. Through a comprehensive empirical evaluation and a case study on the Law School Admissions problem, we demonstrate the effectiveness of ManyFairHPO in balancing multiple fairness objectives, mitigating risks such as self-fulfilling prophecies, and providing interpretable insights to guide stakeholders in making fairness-aware modeling decisions.

Machine learning algorithms use the past and the present to predict the future. But when given biased historical data, these algorithms can quickly become discriminatory. The area of machine learning fairness has emerged to detect and de-bias these algorithms, but has received widespread criticism for its one-size-fits-all approach, which allows certain cases of bias to slip through the cracks. In this study, we take a deeper look at the mechanisms by which machine learning algorithms develop harmful bias. We introduce a new method to interpret discriminatory systems, an Evolutionary algorithm for Feature Interaction (EFI), which we apply to several commonly used machine learning algorithms in two real-world problem instances: violent crime and median house price prediction. In the results, we discover several complex forms of bias including the encoding of race through other seemingly unrelated attributes. Ultimately we suggest that more informative interpretation tools such as EFI can be used to not only explain machine learning outcomes, but supplement and improve existing machine bias detection approaches to provide a more robust and in-depth ethical evaluation of machine learning algorithms.