MARLISA: Multi-Agent Reinforcement Learning with Iterative Sequential Action Selection for Load Shaping of Grid-Interactive Connected Buildings.
José R. Vázquez-CanteliGregor HenzeZoltán NagyPublished in: BuildSys@SenSys (2020)
Keyphrases
- action selection
- multi agent reinforcement learning
- reinforcement learning
- robot soccer
- basal ganglia
- multi agent
- temporal difference
- decision making
- learning agents
- learning agent
- action selection mechanism
- multi agent systems
- function approximation
- state space
- mobile robot
- learning automata
- learning process
- action space
- expert systems
- multi agent learning
- distributed control
- knowledge base
- artificial intelligence
- learning algorithm