Balancing the Scales: Reinforcement Learning for Fair Classification

TL;DR

This paper explores using reinforcement learning, specifically contextual multi-armed bandits, to mitigate bias in imbalanced classification tasks by adapting reward functions for fairness.

Contribution

It introduces a novel RL-based approach employing bandit algorithms to address bias in classification, shifting from traditional representation manipulation methods.

Findings

RL can effectively reduce bias in imbalanced datasets

Adapting reward functions improves fairness outcomes

The approach outperforms some existing bias mitigation techniques

Abstract

Fairness in classification tasks has traditionally focused on bias removal from neural representations, but recent trends favor algorithmic methods that embed fairness into the training process. These methods steer models towards fair performance, preventing potential elimination of valuable information that arises from representation manipulation. Reinforcement Learning (RL), with its capacity for learning through interaction and adjusting reward functions to encourage desired behaviors, emerges as a promising tool in this domain. In this paper, we explore the usage of RL to address bias in imbalanced classification by scaling the reward function to mitigate bias. We employ the contextual multi-armed bandit framework and adapt three popular RL algorithms to suit our objectives, demonstrating a novel approach to mitigating bias.