Determination and optimization of reinforcement learning parameters for driver actions in traffic.
Linsen ChongMontasir M. AbbasBryan HiggsAlejandra MedinaC. Y. David YangPublished in: ITSC (2011)
Keyphrases
- reinforcement learning
- parameter optimization
- maximum likelihood
- action selection
- fine tuning
- markov decision processes
- perceptual aliasing
- parameter determination
- partially observable
- optimization method
- optimal policy
- optimization algorithm
- parameter estimation
- traffic control
- multiagent reinforcement learning
- state and action spaces
- function approximation
- traffic flow
- optimization problems
- state space
- highly non linear
- multi agent
- real time