Termsslug: meta-reinforcement-learning

Meta-Reinforcement Learning

元强化学习

Learning paradigm enabling agents to quickly adapt to new tasks

Curated Intelligence

2 items · Page 1/1
量子位 (微信公众号)T2·Models
62 pts
ICRPM Predicts Task Risk In-Context to Cut Rollout Costs

Researchers from NUDT and Tsinghua propose ICRPM, an in-context risk predictive model that uses recent task-risk observations to forecast risk after the next policy update, avoiding extra rollouts or environment interaction. It improves task risk prediction and sampling in LLM RLVR, Meta-RL and robust control at roughly the same training cost, with an analysis of learning-rate-related first-order lag error.

OpenAI 官方博客T1·Models
15 pts
Meta-RL Exploration Study Title Only

The entry contains only a title regarding meta-reinforcement learning for exploration, lacking body text, specific methods, or experimental results. Technical contribution and value cannot be assessed.