On Fairness, Diversity and Randomness in Algorithmic Decision Making

By Nina Grgic-Hlaca, Muhammad Bilal Zafar, Krishna P. Gummadi, Adrian Weller. 2017

44 citations recorded by the source corpus.

Consider a binary decision making process where a single machine learning classifier replaces a multitude of humans. We raise questions about the resulting loss of diversity in the decision making process. We study the potential benefits of using random classifier ensembles instead of a single classifier in the context of fairness-aware learning and demonstrate various attractive properties: (i) an ensemble of fair classifiers is guaranteed to be fair, for several different measures of fairness, (ii) an ensemble of unfair classifiers can still achieve fair outcomes, and (iii) an ensemble of classifiers can achieve better accuracy-fairness trade-offs than a single classifier. Finally, we introduce notions of distributional fairness to characterize further potential benefits of random classifier ensembles.

Subfields: Cognitive Neuroscience, Safety Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Safety Research

Citations in the corpus

On the Fairness of Causal Algorithmic Recourse (2020, background): “…interventions with minimum quantum size and a fixed budget, it makes sense to spread interventions across a randomly chosen subset since it is not possible to give everyone a very small amount, see (Grgić-Hlača et al. 2017) for broader comments on the potential benefits of randomness in fairness.”
Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer (2017, background): “[16] discuss fairness that can arise out of a mixture of classifiers.”
Fairness in Machine Learning (2020, methodology): “…via the method of Lagrange multipliers [3, 15, 34, 37, 38, 59, 97, 135, 183, 185] or add penalties to the objective [5, 14, 22, 44, 46, 56, 58, 62, 73, 74, 75, 79, 82, 87, 88, 89, 93, 94, 95, 98, 104, 115, 119, 123, 133, 134, 135, 138, 139, 154, 165, 176, 179, 180, 182], others use adversary…”
Transparency: Motivations and Challenges (2017, background): “l. Now consider if the privileged few were chosen uniformly at random { perhaps that might be fair? Notions of fairness beyond equality, and the role of randomness in fairness, were recently explored [24,74]. 4.2 Algorithmic Trading Algorithmic trading is one area where self-interested agents compet”
Predict Responsibly: Increasing Fairness by Learning To Defer (2017, background): “Grgi-Hlaca et al. (2017) discuss fairness that can arise out of a mixture of classifiers.”
FAIR-Ensemble: When Fairness Naturally Emerges From Deep Ensembling (2023, background): “(Grgić-Hlača et al., 2017) points out the theoretical possibility of us-ing an ensemble of randomly selected candidate models to improve fairness, however no empirical validation was presented.”
Transparency: Motivations and Challenges (2019, background): “Now consider if the privileged few were chosen uniformly at random – perhaps that might be fair? Notions of fairness beyond equality, and the role of randomness in fairness, were recently explored [24, 74].”