Pseudorehearsal in actor-critic agents
This addresses forgetting in RL for agents in simple tasks, but appears incremental as it applies known pseudorehearsal to a specific scenario.
The study tackled catastrophic forgetting in reinforcement learning by investigating pseudorehearsal in an actor-critic agent for a pole balancing task, expecting improved performance with proper parameter initialization.
Catastrophic forgetting has a serious impact in reinforcement learning, as the data distribution is generally sparse and non-stationary over time. The purpose of this study is to investigate whether pseudorehearsal can increase performance of an actor-critic agent with neural-network based policy selection and function approximation in a pole balancing task and compare different pseudorehearsal approaches. We expect that pseudorehearsal assists learning even in such very simple problems, given proper initialization of the rehearsal parameters.