Abstract: Preference-based reinforcement learning (PbRL) is a suitable approach for style adaptation of pre-trained robotic behavior: adapting the robot's policy to follow human user preferences while ...
一些您可能无法访问的结果已被隐去。
显示无法访问的结果一些您可能无法访问的结果已被隐去。
显示无法访问的结果