Parameter-Based Value Functions
This addresses a limitation in RL for improving policy evaluation and learning efficiency, though it appears incremental as it builds on existing actor-critic frameworks.
The paper tackles the problem of value functions forgetting useful information about old policies in off-policy actor-critic RL by introducing Parameter-Based Value Functions (PBVFs), which generalize across policies and enable zero-shot learning of new policies that outperform those seen during training, with performance comparable to state-of-the-art methods on control tasks.
Traditional off-policy actor-critic Reinforcement Learning (RL) algorithms learn value functions of a single target policy. However, when value functions are updated to track the learned policy, they forget potentially useful information about old policies. We introduce a class of value functions called Parameter-Based Value Functions (PBVFs) whose inputs include the policy parameters. They can generalize across different policies. PBVFs can evaluate the performance of any policy given a state, a state-action pair, or a distribution over the RL agent's initial states. First we show how PBVFs yield novel off-policy policy gradient theorems. Then we derive off-policy actor-critic algorithms based on PBVFs trained by Monte Carlo or Temporal Difference methods. We show how learned PBVFs can zero-shot learn new policies that outperform any policy seen during training. Finally our algorithms are evaluated on a selection of discrete and continuous control tasks using shallow policies and deep neural networks. Their performance is comparable to state-of-the-art methods.