Mean-Variance Policy Iteration for Risk-Averse Reinforcement Learning
This work addresses risk-sensitive control for applications like robotics, though it is incremental by building on existing methods like TD3.
The paper tackles risk-averse reinforcement learning by proposing a mean-variance policy iteration (MVPI) framework that optimizes reward variance in MDPs, resulting in a risk-averse TD3 method that outperforms vanilla TD3 and previous methods in Mujoco robot simulations under a risk-aware metric.
We present a mean-variance policy iteration (MVPI) framework for risk-averse control in a discounted infinite horizon MDP optimizing the variance of a per-step reward random variable. MVPI enjoys great flexibility in that any policy evaluation method and risk-neutral control method can be dropped in for risk-averse control off the shelf, in both on- and off-policy settings. This flexibility reduces the gap between risk-neutral control and risk-averse control and is achieved by working on a novel augmented MDP directly. We propose risk-averse TD3 as an example instantiating MVPI, which outperforms vanilla TD3 and many previous risk-averse control methods in challenging Mujoco robot simulation tasks under a risk-aware performance metric. This risk-averse TD3 is the first to introduce deterministic policies and off-policy learning into risk-averse reinforcement learning, both of which are key to the performance boost we show in Mujoco domains.