Generalized Population-Based Training for Hyperparameter Optimization in Reinforcement Learning

arXiv:2404.08233v1 Announce Type: new
Abstract: Hyperparameter optimization plays a key role in the machine learning domain. Its significance is especially pronounced in reinforcement learning (RL), where agents continuously interact with and adapt to their environments, requiring dynamic adjustments in their learning trajectories. To cater to this dynamicity, the Population-Based Training (PBT) was introduced, leveraging the collective intelligence of a population of agents learning simultaneously. However, PBT tends to favor high-performing agents, potentially neglecting the explorative potential of agents on the …

abstract adapt agents arxiv cs.ai cs.lg cs.ne domain dynamic environments generalized hyperparameter key machine machine learning optimization population reinforcement reinforcement learning role significance training type