论文官方一手00:17UCB探索：Q集成方法对RL从业者而言，Q-ensembles提供了一种实用且可扩展的UCB探索方案，有助于提升复杂任务的学习效率。#reinforcement-learning#exploration#ucb#q-ensemblesO官方一手OpenAI Blog原文稍后读已读值得跟进有用关注 reinforcement-learning