Risk-Sensitive Reinforcement Learning via Policy Gradient Search
This is an incremental survey paper that addresses the problem of incorporating risk into RL for practitioners and researchers, without presenting new experimental results.
The paper surveys risk-sensitive reinforcement learning (RL) approaches that incorporate risk measures into policy optimization, covering settings where risk is a constraint or the objective, and focuses on policy gradient methods as the solution approach.
The objective in a traditional reinforcement learning (RL) problem is to find a policy that optimizes the expected value of a performance metric such as the infinite-horizon cumulative discounted or long-run average cost/reward. In practice, optimizing the expected value alone may not be satisfactory, in that it may be desirable to incorporate the notion of risk into the optimization problem formulation, either in the objective or as a constraint. Various risk measures have been proposed in the literature, e.g., exponential utility, variance, percentile performance, chance constraints, value at risk (quantile), conditional value-at-risk, prospect theory and its later enhancement, cumulative prospect theory. In this book, we consider risk-sensitive RL in two settings: one where the goal is to find a policy that optimizes the usual expected value objective while ensuring that a risk constraint is satisfied, and the other where the risk measure is the objective. We survey some of the recent work in this area specifically where policy gradient search is the solution approach. In the first risk-sensitive RL setting, we cover popular risk measures based on variance, conditional value-at-risk, and chance constraints, and present a template for policy gradient-based risk-sensitive RL algorithms using a Lagrangian formulation. For the setting where risk is incorporated directly into the objective function, we consider an exponential utility formulation, cumulative prospect theory, and coherent risk measures. This non-exhaustive survey aims to give a flavor of the challenges involved in solving risk-sensitive RL problems using policy gradient methods, as well as outlining some potential future research directions.