Improving quality of prediction in highly dynamic environments using approximate dynamic programming.
Rajesh GanesanPoornima BalakrishnaLance SherryPublished in: Qual. Reliab. Eng. Int. (2010)
Keyphrases
- approximate dynamic programming
- highly dynamic environments
- linear program
- reinforcement learning
- stochastic dynamic programming
- dynamic programming
- control policy
- data mining
- factored mdps
- special case
- probability distribution
- markov random field
- learning tasks
- function approximation
- step size
- multiscale
- learning algorithm