Rethinking Exploration in Reinforcement Learning with Effective Metric-Based Exploration Bonus Yiming Wang 1

Enhancing exploration in reinforcement learning (RL) through the incorporation of intrinsic rewards, specifically by leveraging state discrepancy measures within various metric spaces as exploration bonuses, has emerged as a prevalent strategy to encourage agents to visit novel states. The critical factor lies in how to quantify the difference between adjacent states as novelty for promoting effective exploration.