Interval-Based Markov Decision Processes for Regulating Interactions Between Two Agents in Multi-agent Systems.
Graçaliz Pereira DimuroAntônio Carlos da Rocha CostaPublished in: PARA (2004)
Keyphrases
- markov decision processes
- multi agent systems
- multi agent
- finite state
- optimal policy
- transition matrices
- agent systems
- state space
- reinforcement learning
- autonomous agents
- dynamic programming
- decentralized control
- single agent
- cooperative
- software agents
- coalition formation
- reachability analysis
- decision theoretic planning
- agent architecture
- agent technology
- planning under uncertainty
- policy iteration
- autonomous entities
- finite horizon
- action space
- temporal reasoning
- multiagent systems
- average cost
- real valued
- game theory
- stochastic games
- average reward
- reinforcement learning algorithms
- decision processes
- model based reinforcement learning
- partially observable
- trust model
- state and action spaces
- multiple agents
- infinite horizon
- expected reward
- action sets
- search space
- multiagent reinforcement learning
- state abstraction
- markov decision problems
- markov decision process