Multiagent Learning for Black Box System Reward Functions.
Kagan TumerAdrian K. AgoginoPublished in: Adv. Complex Syst. (2009)
Keyphrases
- black box
- multiagent learning
- reward function
- reinforcement learning
- learning agents
- multiple agents
- multi agent
- reinforcement learning algorithms
- markov decision processes
- single agent
- state space
- optimal policy
- partially observable
- black boxes
- white box
- resource allocation
- multiagent systems
- transition probabilities
- multi agent learning
- markov decision process
- multi agent systems
- learning agent
- state variables
- game theoretic
- model free
- integration testing
- average reward
- white box testing
- state transition
- test cases
- dynamic programming
- prior knowledge
- bayesian networks
- nash equilibria
- sufficient conditions
- markov chain
- stochastic games
- source code
- training data
- learning algorithm
- machine learning