4 papers
LLM-Guided Communication for Cooperative Multi-Agent Reinforcement Learning
Sangjun Bae, Yisak Park, Sanghyeon Lee +1
Communication is a key component in multi-agent reinforcement learning (MARL) for mitigating partial observability, yet prior approaches often rely on inefficient information excha…
Strict Subgoal Execution: Reliable Long-Horizon Planning in Hierarchical Reinforcement Learning
Jaebak Hwang, Sanghyeon Lee, Jeongmo Kim +1
Long-horizon goal-conditioned tasks pose fundamental challenges for reinforcement learning (RL), particularly when goals are distant and rewards are sparse. While hierarchical and…
Self-Improving Skill Learning for Robust Skill-based Meta-Reinforcement Learning
Sanghyeon Lee, Sangjun Bae, Yisak Park +1
Meta-reinforcement learning (Meta-RL) facilitates rapid adaptation to unseen tasks but faces challenges in long-horizon environments. Skill-based approaches tackle this by decompos…
Exclusively Penalized Q-learning for Offline Reinforcement Learning
Junghyuk Yeom, Yonghyeon Jo, Jungmo Kim +2
Constraint-based offline reinforcement learning (RL) involves policy constraints or imposing penalties on the value function to mitigate overestimation errors caused by distributio…