Applying Experimental Design and Regression Splines to High-Dimensional Continuous-State Stochastic Dynamic Programming.
Victoria C. P. ChenDavid RuppertChristine A. ShoemakerPublished in: Oper. Res. (1999)
Keyphrases
- stochastic dynamic programming
- continuous state
- experimental design
- high dimensional
- reinforcement learning
- robot navigation
- finite state
- control policies
- state dependent
- active learning
- action space
- planning problems
- partially observable markov decision processes
- dimensionality reduction
- model selection
- approximate dynamic programming
- empirical studies
- feature selection
- feature space
- high dimensionality
- markov decision processes
- domain independent
- virtual learning environments
- sample size