AI RESEARCH
Accelerating Robotic Reinforcement Learning with Agent Guidance
arXiv CS.AI
•
ArXi:2602.11978v2 Announce Type: replace-cross Reinforcement Learning (RL) offers a powerful paradigm for autonomous robots to master generalist manipulation skills through trial-and-error. However, its real-world application is stifled by low sample efficiency. Recent Human-in-the-Loop (HIL) methods accelerate