PapersBuilders73°11:41Chinese contentCAST结合规划器引导行为提升强化学习新提出的CAST方法结合规划器与策略,在强化学习中实现更高效的值学习,并在四足机器人上验证效果。#CAST#强化学习#模型基础#规划器aarXiv: Google DeepMind@Pietro Noah Crestaz et al. (4)SourceRead laterReadWorth followingUsefulFollow CAST
How-toBuildersMulti-source18:34Chinese contentAndrej Karpathy:先掌握模型再谈智能体,否则代价惨重Karpathy用自己的失败经历点醒你:别急着搞智能体,先把模型弄透。少走5年弯路。#Andrej Karpathy#OpenAI#智能体#模型基础10 个信源在谈EventAI Will@FinanceYF511 sourcesSourceRead laterReadWorth followingUsefulFollow Andrej Karpathy