Multi-task Learning and Catastrophic Forgetting in Continual Reinforcement Learning
This addresses the problem of catastrophic forgetting in continual reinforcement learning for AI agents, but it is incremental as it builds on existing methods like EWC and multi-task learning.
The paper investigates whether multi-task deep reinforcement learning can outperform single-task methods on a new similar task and whether elastic weight consolidation (EWC) helps retain performance on new tasks while reducing catastrophic forgetting on previous ones. Results show that a multi-task GA3C algorithm outperforms single-task versions on a new task (Phoenix) and that EWC-augmented multi-task GA3C achieves similar new-task performance while mitigating forgetting on prior tasks (Space Invaders and Demon Attack).
In this paper we investigate two hypothesis regarding the use of deep reinforcement learning in multiple tasks. The first hypothesis is driven by the question of whether a deep reinforcement learning algorithm, trained on two similar tasks, is able to outperform two single-task, individually trained algorithms, by more efficiently learning a new, similar task, that none of the three algorithms has encountered before. The second hypothesis is driven by the question of whether the same multi-task deep RL algorithm, trained on two similar tasks and augmented with elastic weight consolidation (EWC), is able to retain similar performance on the new task, as a similar algorithm without EWC, whilst being able to overcome catastrophic forgetting in the two previous tasks. We show that a multi-task Asynchronous Advantage Actor-Critic (GA3C) algorithm, trained on Space Invaders and Demon Attack, is in fact able to outperform two single-tasks GA3C versions, trained individually for each single-task, when evaluated on a new, third task, namely, Phoenix. We also show that, when training two trained multi-task GA3C algorithms on the third task, if one is augmented with EWC, it is not only able to achieve similar performance on the new task, but also capable of overcoming a substantial amount of catastrophic forgetting on the two previous tasks.