BriefGPT.xyz
Dec, 2023
基于语言条件的语义搜索导向的机器人操作任务策略
Language-Conditioned Semantic Search-Based Policy for Robotic Manipulation Tasks
HTML
PDF
Jannik Sheikh, Andrew Melnik, Gora Chand Nandi, Robert Haschke
TL;DR
我们提出了一种基于语言条件的语义搜索方法,从可用的示范数据集中获取在线搜索策略,在CALVIN基准测试中超过了基线性能,并展现了强大的零样本适应能力,对于扩展基于在线搜索策略的任务具有巨大的潜力。
Abstract
reinforcement learning
and
imitation learning
approaches utilize
policy learning
strategies that are difficult to generalize well with jus
→