Awesome Robotics Manipulation · full_paper

SafeMIL: Learning Offline Safe Imitation Policy from Non-Preferred Trajectories

作者：Returaj Burnwal, Nirav Pravinbhai Bhatt, Balaraman Ravindran · 单位：Department of Computer Science and Engineering · 会议/期刊：AAAI 2026 · 日期：2025-11-11 · 来源：Low-Level Learning-Based Action Modelling / Learning Strategy / Imitation Learning from Action - Efficiency and Robustness / Robustness

模仿学习安全机器人学习
新标签打开 AlphaXiv新标签打开 Paper
SafeMIL: Learning Offline Safe Imitation Policy from Non-Preferred Trajectories figure
AlphaXiv 中文论文页面（可滚动查看）