📡 Embodied AI Radar具身智能雷达

Topics主题 › Robot Learning: RL & Imitation机器人学习:强化学习与模仿学习

Paper论文 2021 · Google CoRL 2021

Implicit Behavioral Cloning

Represents the policy as an energy-based model over actions instead of a regressor, which handles multimodal and discontinuous demonstrations much better.用基于能量的模型(而非直接回归)表示策略,更好地处理多模态、不连续的示范数据。

BibTeX
@misc{florence2021implicit,
  title = {{Implicit Behavioral Cloning}},
  author = {Pete Florence and Corey Lynch and Andy Zeng and Oscar Ramirez and Ayzaan Wahid and Laura Downs and Adrian Wong and Johnny Lee and Igor Mordatch and Jonathan Tompson},
  year = {2021},
  eprint = {2109.00137},
  archivePrefix = {arXiv},
  primaryClass = {cs.RO},
  url = {https://arxiv.org/abs/2109.00137},
}

Authors (from arXiv): 作者(来自 arXiv):Pete Florence, Corey Lynch, Andy Zeng, Oscar Ramirez, Ayzaan Wahid, Laura Downs, Adrian Wong, Johnny Lee et al.

More from: 同机构更多条目:Google DeepMind / Google谷歌 DeepMind / 谷歌

More in Robot Learning: RL & Imitation更多机器人学习:强化学习与模仿学习条目

All Robot Learning: RL & Imitation entries →查看全部机器人学习:强化学习与模仿学习条目 →