Solving F3MDPs: Collaborative Multiagent Markov Decision Processes with Factored Transitions, Rewards and Stochastic Policies.
Julia RadoszyckiNathalie PeyrardRégis SabbadinPublished in: PRIMA (2015)
Keyphrases
- markov decision processes
- policy iteration algorithm
- factored mdps
- algebraic decision diagrams
- state space
- optimal policy
- planning under uncertainty
- multi agent
- markov decision problems
- finite state
- reinforcement learning
- stochastic games
- policy iteration
- reward function
- fully observable
- transition matrices
- dynamic programming
- control policies
- semi markov decision processes
- average cost
- markov decision process
- total reward
- decision processes
- partially observable
- average reward
- decision problems
- reachability analysis
- finite horizon
- markov games
- infinite horizon
- discounted reward
- sequential decision making under uncertainty
- decision theoretic planning
- partially observable markov decision processes
- reinforcement learning algorithms
- decentralized control
- state and action spaces
- expected reward
- dec pomdps
- model based reinforcement learning
- multiagent systems
- action space
- stationary policies
- discount factor
- policy evaluation
- probabilistic planning
- real time dynamic programming
- stochastic shortest path
- continuous state spaces
- action sets