Subequivariant Reinforcement Learning in 3D Multi-Entity Physical Environments.
Runfa ChenLing WangYu DuTianrui XueFuchun SunJianwei ZhangWenbing HuangPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- physical world
- real world
- physical objects
- physical environment
- function approximation
- optimal policy
- multi agent
- state space
- coreference resolution
- complex environments
- dynamic programming
- learning process
- learning algorithm
- machine learning
- physical space
- policy search
- multi agent environments
- robot control
- entity identification
- reinforcement learning algorithms
- action selection
- autonomous robots
- evolutionary algorithm
- case study
- knowledge base
- data mining