Enhancing Group Fairness in Online Settings Using Oblique Decision Forests
Somnath Basu Roy Chowdhury, Nicholas Monath, Ahmad Beirami, Rahul Kidambi, Avinava Dubey, Amr Ahmed, Snigdha Chaturvedi
TL;DR
The paper targets online group fairness, addressing the difficulty of enforcing demographic parity when data arrives one sample at a time. It introduces Aranyani, an ensemble of soft-routed oblique decision trees that isolate node-level parameters and use aggregate statistics to estimate fairness-gradients online without storing past data. The authors provide a gradient-based training framework with a Hubera-regularized node-level constraint, plus theoretical guarantees on DP bounds, gradient estimation, and convergence. Experiments across tabular, vision, and language benchmarks show Aranyani achieves superior accuracy-fairness trade-offs and notable efficiency gains over baselines, highlighting its practical potential for online, fairness-aware decision systems.
Abstract
Fairness, especially group fairness, is an important consideration in the context of machine learning systems. The most commonly adopted group fairness-enhancing techniques are in-processing methods that rely on a mixture of a fairness objective (e.g., demographic parity) and a task-specific objective (e.g., cross-entropy) during the training process. However, when data arrives in an online fashion -- one instance at a time -- optimizing such fairness objectives poses several challenges. In particular, group fairness objectives are defined using expectations of predictions across different demographic groups. In the online setting, where the algorithm has access to a single instance at a time, estimating the group fairness objective requires additional storage and significantly more computation (e.g., forward/backward passes) than the task-specific objective at every time step. In this paper, we propose Aranyani, an ensemble of oblique decision trees, to make fair decisions in online settings. The hierarchical tree structure of Aranyani enables parameter isolation and allows us to efficiently compute the fairness gradients using aggregate statistics of previous decisions, eliminating the need for additional storage and forward/backward passes. We also present an efficient framework to train Aranyani and theoretically analyze several of its properties. We conduct empirical evaluations on 5 publicly available benchmarks (including vision and language datasets) to show that Aranyani achieves a better accuracy-fairness trade-off compared to baseline approaches.