Adaptive value function approximation for continuous-state stochastic dynamic programming.
Huiyuan FanPrashant K. TarunVictoria C. P. ChenPublished in: Comput. Oper. Res. (2013)
Keyphrases
- continuous state
- stochastic dynamic programming
- reinforcement learning
- robot navigation
- finite state
- approximate dynamic programming
- action space
- state action
- control policies
- state dependent
- planning problems
- partially observable markov decision processes
- state space
- evaluation function
- temporal difference
- single agent
- hidden state
- experimental design
- optimal policy