Useful Policy Invariant Shaping from Arbitrary Advice
This work addresses the problem of reducing data requirements in reinforcement learning for researchers and practitioners, but it is incremental as it builds on existing reward shaping methods.
The paper identifies a flaw in the dynamic potential based advice (DPBA) method for reward shaping in reinforcement learning, which aims to improve learning efficiency without altering the optimal policy, and proposes a new method called policy invariant explicit shaping (PIES) that successfully addresses this flaw, as demonstrated theoretically and empirically.
Reinforcement learning is a powerful learning paradigm in which agents can learn to maximize sparse and delayed reward signals. Although RL has had many impressive successes in complex domains, learning can take hours, days, or even years of training data. A major challenge of contemporary RL research is to discover how to learn with less data. Previous work has shown that domain information can be successfully used to shape the reward; by adding additional reward information, the agent can learn with much less data. Furthermore, if the reward is constructed from a potential function, the optimal policy is guaranteed to be unaltered. While such potential-based reward shaping (PBRS) holds promise, it is limited by the need for a well-defined potential function. Ideally, we would like to be able to take arbitrary advice from a human or other agent and improve performance without affecting the optimal policy. The recently introduced dynamic potential based advice (DPBA) method tackles this challenge by admitting arbitrary advice from a human or other agent and improves performance without affecting the optimal policy. The main contribution of this paper is to expose, theoretically and empirically, a flaw in DPBA. Alternatively, to achieve the ideal goals, we present a simple method called policy invariant explicit shaping (PIES) and show theoretically and empirically that PIES succeeds where DPBA fails.