Adaptive Action Supervision in Reinforcement Learning from Real-World Multi-Agent Demonstrations.
Keisuke FujiiKazushi TsutsuiAtom ScottHiroshi NakaharaNaoya TakeishiYoshinobu KawaharaPublished in: ICAART (2) (2024)
Keyphrases
- reinforcement learning
- multi agent
- real world
- action selection
- multi agent systems
- state action
- action space
- function approximation
- partially observable domains
- state space
- cooperative
- wide range
- synthetic data
- autonomous agents
- single agent
- data sets
- reinforcement learning agents
- multi agent reinforcement learning
- learning agents
- learning capabilities
- adaptive control
- markov decision processes
- learning algorithm
- agent oriented
- machine learning
- model free
- case study
- transition model
- reinforcement learning algorithms
- data mining
- continuous state
- agent learns
- reward shaping
- active learning