Selecting GVT interval for time-warp-based distributed simulation using reinforcement learning technique.
Jun WangCarl TropperPublished in: SpringSim (2009)
Keyphrases
- reinforcement learning
- multi agent
- distributed systems
- distributed environment
- lightweight
- peer to peer
- simulation model
- function approximation
- distributed data
- state space
- real time
- cooperative
- machine learning
- real robot
- communication cost
- computing environments
- robocup soccer
- fault tolerant
- markov decision processes
- transfer learning
- mobile agents
- optimal policy
- learning process