Novelty Detection in Reinforcement Learning with World Models

arXiv:2310.08731v2 Announce Type: replace-cross
Abstract: Reinforcement learning (RL) using world models has found significant recent successes. However, when a sudden change to world mechanics or properties occurs then agent performance and reliability can dramatically decline. We refer to the sudden change in visual properties or state transitions as novelties. Implementing novelty detection within generated world model frameworks is a crucial task for protecting the agent when deployed. In this paper, we propose straightforward bounding approaches to incorporate novelty detection into …

abstract agent arxiv change cs.ai cs.lg cs.sy detection eess.sy found however performance reinforcement reinforcement learning reliability state transitions type visual world world models