| 1 |
Language-Conditioned Offline RL for Multi-Robot Navigation |
提出一种基于离线强化学习和语言模型的用于多机器人导航的策略学习方法 |
reinforcement learning offline RL offline reinforcement learning |
|
|
| 2 |
Theia: Distilling Diverse Vision Foundation Models for Robot Learning |
Theia：为机器人学习提炼多样化视觉基础模型，提升泛化能力 |
policy learning foundation model |
|
|
| 3 |
A Differential Dynamic Programming Framework for Inverse Reinforcement Learning |
提出基于DDP的逆强化学习框架，用于从演示中恢复成本函数、系统动力学和约束。 |
reinforcement learning inverse reinforcement learning |
|
|
| 4 |
Privileged Reinforcement and Communication Learning for Distributed, Bandwidth-limited Multi-robot Exploration |
提出基于特权强化学习和通信学习的分布式多机器人探索方法，解决带宽限制问题。 |
reinforcement learning deep reinforcement learning |
|
|
| 5 |
Detecting Unsafe Behavior in Neural Network Imitation Policies for Caregiving Robotics |
针对照护机器人，提出基于集成预测器和归一化流的异常检测方法，提升模仿学习策略的安全性。 |
imitation learning diffusion policy |
|
|