On opportunistic techniques for solving decentralized Markov decision processes with temporal constraints.
Janusz MareckiMilind TambePublished in: AAMAS (2007)
Keyphrases
- markov decision processes
- temporal constraints
- transition matrices
- semi markov decision processes
- temporal reasoning
- decentralized control
- simple temporal
- finite state
- optimal policy
- state space
- reinforcement learning
- dec mdps
- markov decision problems
- action sets
- policy iteration
- temporal information
- decision theoretic planning
- dynamic programming
- average reward
- planning under uncertainty
- partially observable
- average cost
- infinite horizon
- factored mdps
- stochastic shortest path
- linear programming
- markov decision process
- distributed systems
- temporal planning
- qualitative and quantitative
- real time dynamic programming
- lower bound
- multi agent