Fairness-Aware Active Learning for Decoupled Model

Yiting Cao, Chao Lan

Abstract: Active learning (AL) aims to actively label training data for supervised learners to efficiently improve accuracy of the trained models. However, traditional AL techniques ignore model fairness, which is an important social norm in modern machine learning applications. In this paper, we propose the first fairness-aware active learning technique for decoupled model called D-FA <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> L. It labels data that receive disagreed predictions from the decoupled models, and we hypothesize it can simultaneously improve model fairness and model accuracy. We present theoretical justification of D-FA <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> L, proving it can reduce model unfairness under proper conditions. In experiments, we show D-FA <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> L can reduce model unfairness effectively, and the reduction is more efficient than both classic and state-of-the-art AL techniques on three data sets. In the meantime, D-FA <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> L is able to maintain similar improvement rate on model accuracy.