Learning fair policies in multi-objective (deep) reinforcement learning with average and discounted rewards

As the operations of autonomous systems generally affect simultaneously several users, it is crucial that their designs account for fairness considerations. In contrast to standard (deep) reinforcement learning (RL), in this paper, we investigate the problem of learning a policy that treats its users equitably. In this paper, we formulate this novel RL problem, in which an objective function, which encodes a notion of fairness that we formally define, is optimized. For this problem, we provide a theoretical discussion where we examine the case of discounted rewards and that of average rewards. During this analysis, we notably derive a new result in the standard RL setting, which is of independent interest: it states a novel bound on the approximation error with respect to the optimal average reward of that of a policy optimal for the discounted reward. Since learning with discounted rewards is generally easier, this discussion further justifies finding a fair policy for the average reward by learning a fair policy for the discounted reward. We propose three novel deep RL adaptations to learn fair policies. We evaluate these methods on a set of fair multi-objective deep RL benchmarks, where we show that they outperform the state-of-the-art methods in terms of fairness and performance. Download paper here

Recommended citation: Siddique, Umer, Paul Weng, and Matthieu Zimmer. "Learning fair policies in multi-objective (deep) reinforcement learning with average and discounted rewards." International Conference on Machine Learning. PMLR, 2020.
Download Paper | Download Slides