Representativity Fairness in Clustering

Abstract

Incorporating fairness constructs into machine learning algorithms is a topic of much societal importance and recent interest. Clustering, a fundamental task in unsupervised learning that manifests across a number of web data scenarios, has also been subject of attention within fair ML research. In this paper, we develop a novel notion of fairness in clustering, called representativity fairness. Representativity fairness is motivated by the need to alleviate disparity across objects' proximity to their assigned cluster representatives, to aid fairer decision making. We illustrate the importance of representativity fairness in real-world decision making scenarios involving clustering and provide ways of quantifying objects' representativity and fairness over it. We develop a new clustering formulation, RFKM, that targets to optimize for representativity fairness along with clustering quality. Inspired by the-Means framework, RFKM incorporates novel loss terms to formulate an objective function. The RFKM objective and optimization approach guides it towards clustering configurations that yield higher representativity fairness. Through an empirical evaluation over a variety of public datasets, we establish the effectiveness of our method. We illustrate that we are able to significantly improve representativity fairness at only marginal impact to clustering quality.

A guide to formulating fairness in an optimization model

“…Counterfactual fairness (Kusner et al, 2017;Russell et al, 2017) is another useful scheme that aims to avoid undesirable confounding factors for evaluating fairness in classification, and it is unclear at this point how to incorporate this scheme into an optimization model. Beyond fairness in classification, recent research has also seen progress on fairness in unsupervised learning (e.g., Abraham et al, 2019, Deepak & Abraham, 2020 and reinforcement learning (e.g., Weng, 2019, Siddique et al, 2020. We do not discuss these machine learning frameworks, as they are difficult to interpret as optimization models even without any fairness consideration.…”

AI and core electoral processes: Mapping the horizons

“…As may be obvious, minimizing the cumulative distance could still leave some voters with a significant distance to travel, to reach their polling booth. Recent advances in fair clustering (P and Abraham 2020; Jung, Kannan, and Lutz 2020; Abbasi, Bhaskara, and Venkatasubramanian 2021; Stepanov 2022) consider ensuring that no objects (voters) are left too disadvantaged in terms of distance to their cluster representative (polling booth). This and other advances in fair clustering (Mahabadi and Vakilian 2020; Kleindessner, Awasthi, and Morgenstern 2020) have extended the clustering literature to align more with considerations that are applicable for polling booth location determination.…”

Exploring Rawlsian Fairness for K-Means Clustering

“…Davidson and Ravi [4] look at postprocessing clusters for fairness; they do so by presenting it as a Minimum Cluster Modification for Group Fairness (MCMF) optimisation problem which is formulated as an ILP. Other notions of fairness such as representativity fairness [13] and proportionality fairness [2] have also been proposed for clustering.…”