Macro-Action-Based Multi-Agent/Robot Deep Reinforcement Learning under Partial Observability.
Yuchen XiaoPublished in: CoRR (2022)
Keyphrases
- partial observability
- reinforcement learning
- multi agent
- multi agent planning
- partially observable
- action selection
- learning agent
- mobile robot
- agent programming
- action space
- symbolic model checking
- function approximation
- state space
- fully observable
- markov decision process
- partially observable markov decision processes
- single agent
- belief space
- multi robot
- markov decision processes
- machine learning
- learning algorithm
- reinforcement learning algorithms
- path planning
- motion planning
- robotic systems
- temporal difference
- humanoid robot
- autonomous agents
- planning under partial observability
- belief state
- human actions
- optimal policy
- dynamic programming
- multi agent systems
- learning capabilities
- initial state
- planning problems
- action sequences
- transfer learning
- action recognition