Proximal Policy Optimization with Adaptive Exploration

arXiv:2405.04664v1 Announce Type: new
Abstract: Proximal Policy Optimization with Adaptive Exploration (axPPO) is introduced as a novel learning algorithm. This paper investigates the exploration-exploitation tradeoff within the context of reinforcement learning and aims to contribute new insights into reinforcement learning algorithm design. The proposed adaptive exploration framework dynamically adjusts the exploration magnitude during training based on the recent performance of the agent. Our proposed method outperforms standard PPO algorithms in learning efficiency, particularly when significant exploratory behavior is needed at …

abstract algorithm algorithm design arxiv context cs.ai cs.lg design exploitation exploration framework insights novel optimization paper policy reinforcement reinforcement learning training type