Awesome RL with Human Feedback / 2020 and before / arXiv

Reward learning from human preferences and demonstrations in Atari

作者：Borja Ibarz, Jan Leike, Tobias Pohlen, Geoffrey Irving, Shane Legg, Dario Amodei · 来源：arXiv · 发布日期：2020

奖励黑客/过优化鲁棒性/噪声反馈奖励建模偏好学习人类反馈设计
新标签打开 AlphaXiv新标签打开 Code
Reward learning from human preferences and demonstrations in Atari figure
AlphaXiv 中文概览（可滚动查看）