Robust Risk-Aware Reinforcement Learning

TL;DR

This paper introduces a risk-aware reinforcement learning framework that optimizes policies considering worst-case model uncertainty using Wasserstein balls, demonstrated on financial tasks.

Contribution

It develops explicit policy gradient methods for robust risk-aware RL under model uncertainty, integrating RDEU for flexible risk-reward profiles.

Findings

Effective in three financial applications

Outperforms non-robust methods in uncertain environments

Provides a new approach to risk-sensitive policy optimization

Abstract

We present a reinforcement learning (RL) approach for robust optimisation of risk-aware performance criteria. To allow agents to express a wide variety of risk-reward profiles, we assess the value of a policy using rank dependent expected utility (RDEU). RDEU allows the agent to seek gains, while simultaneously protecting themselves against downside risk. To robustify optimal policies against model uncertainty, we assess a policy not by its distribution, but rather, by the worst possible distribution that lies within a Wasserstein ball around it. Thus, our problem formulation may be viewed as an actor/agent choosing a policy (the outer problem), and the adversary then acting to worsen the performance of that strategy (the inner problem). We develop explicit policy gradient formulae for the inner and outer problems, and show its efficacy on three prototypical financial problems: robust portfolio allocation, optimising a benchmark, and statistical arbitrage.