A multi-agent prediction market based on partially observable stochastic game.
Janyl JumadinovaPrithviraj DasguptaPublished in: ICEC (2011)
Keyphrases
- partially observable
- multi agent
- reinforcement learning
- decision problems
- state space
- dynamical systems
- markov decision processes
- fully observable
- partial observability
- game theoretic
- game theory
- markov decision problems
- multi agent systems
- belief state
- partially observable domains
- single agent
- action models
- partial observations
- infinite horizon
- optimal strategy
- initially unknown
- partially observable environments
- partially observable markov decision processes
- reward function
- nash equilibrium
- multiple agents
- belief space
- monte carlo
- multiagent systems
- optimal policy
- hidden markov models
- machine learning
- utility function
- coalition formation