Phasic Diversity Optimization for Population-Based Reinforcement Learning

arXiv:2403.11114v1 Announce Type: new
Abstract: Reviewing the previous work of diversity Rein-forcement Learning,diversity is often obtained via an augmented loss function,which requires a balance between reward and diversity.Generally,diversity optimization algorithms use Multi-armed Bandits algorithms to select the coefficient in the pre-defined space. However, the dynamic distribution of reward signals for MABs or the conflict between quality and diversity limits the performance of these methods. We introduce the Phasic Diversity Optimization (PDO) algorithm, a Population-Based Training framework that separates reward and …

abstract algorithms arxiv balance cs.ai cs.lg distribution diversity dynamic function however loss multi-armed bandits optimization population reinforcement reinforcement learning space type via work