SPaRM:an efficient exploration and planning framework for sparse reward reinforcement learning

• Published in: 高技术通讯（英文版） Vol. 30; no. 4; pp. 344 - 355
• Main Authors: BAN Jian, LI Gongyan, XU Shaoyun
• Format: Journal Article
• Language: English
• Published: Institute of Microelectronics,Chinese Academy of Sciences,Beijing 100029,P.R.China 01.12.2024
• Subjects: reinforcement learning(RL); sparse reward; space partitioning(SP); reward-free exploration(RFE); reverse merging(RM)

Due to the issue of long-horizon,a substantial number of visits to the state space is required during the exploration phase of reinforcement learning(RL)to gather valuable information.Addi-tionally,due to the challenge posed by sparse rewards,the planning phase of reinforcement learning consumes a c...