review:2021-06_reinforcement_learning_as_one_big_sequence_modeling_problem

2021-06 Reinforcement Learning as One Big Sequence Modeling Problem

https://arxiv.org/abs/2106.02039

star, transformer, RL, motion, state discretization, action discretization, beam search, autoregression, goal-conditioned RL, offline RL, world model, Michael Janner, Sergey Levine, 2021

review/2021-06_reinforcement_learning_as_one_big_sequence_modeling_problem.txt · 마지막으로 수정됨: 2024/04/25 00:15 저자 rex8312