Max-norm Projections for Factored MDPs.
Carlos GuestrinDaphne KollerRonald ParrPublished in: IJCAI (2001)
Keyphrases
- factored mdps
- context specific
- markov decision processes
- state space
- algebraic decision diagrams
- approximate dynamic programming
- stochastic processes
- markov decision problems
- basis functions
- policy iteration
- reinforcement learning
- finite state machines
- planning under uncertainty
- probability distribution
- general purpose
- decision makers