Continuous-Time Distributed Dynamic Programming for Networked Multi-Agent Markov Decision Processes.
Donghwan LeeHan-Dong LimDo Wan KimPublished in: ICCA (2024)
Keyphrases
- markov decision processes
- dynamic programming
- multi agent
- state space
- reinforcement learning
- optimal control
- optimal policy
- infinite horizon
- finite state
- planning under uncertainty
- cooperative
- decentralized control
- partially observable
- multi agent systems
- reinforcement learning algorithms
- policy iteration
- factored mdps
- markov chain
- decision theoretic planning
- finite horizon
- stationary policies
- average cost
- average reward
- decision processes
- transition matrices
- model based reinforcement learning
- linear programming
- risk sensitive
- multistage
- action sets
- reachability analysis
- markov decision problems
- state and action spaces
- learning algorithm
- partially observable markov decision processes
- multiple agents
- data mining
- action space
- markov decision process
- multiagent systems
- search algorithm
- interval estimation
- stochastic shortest path