Scalable Initial State Interdiction for Factored MDPs.
Swetasudha PandaYevgeniy VorobeychikPublished in: IJCAI (2018)
Keyphrases
- initial state
- factored mdps
- state space
- markov decision processes
- optimal policy
- average cost
- markov decision problems
- context specific
- situation calculus
- approximate dynamic programming
- algebraic decision diagrams
- policy iteration
- markov decision process
- reinforcement learning
- dynamic programming
- heuristic search
- stochastic processes
- probability distribution
- markov chain
- planning problems
- basis functions
- infinite horizon
- search space
- mathematical model
- finite state
- state variables
- multistage
- partially observable
- decision makers
- planning under uncertainty
- dynamical systems