01-09-2014, 01:32 PM
Human mind and computations
Human.pptx (Size: 462.41 KB / Downloads: 10)
Single-agent learning
An agent interacts with the environment by selecting actions to take and then perceiving the effects of those actions,
A new state and a reward signal indicating if it has reached some goal
The objective of the agent is to maximize some measure over the rewards, like the sum of all rewards after a number of actions taken
Multi-Agent Learning :
It is Computerized system composed of multiple interacting intelligent agents within an environment.
Agent tries to solve its learning problem, while other agents in the environment also are trying to solve their own learning problems.
Used to solve problems that are difficult or impossible for an individual agent.
Theoretical solution to fully specified problem: game theory
A multi-agent system may contain combined human-agent teams.
Conclusion
Human mind demonstrates the importance of carefully taking into account the information one learns during the course of play of a game
Computations show how a game-theoretic model (i.e. regular & evolutionary game theory) that incorporates the information and incentives of other participants helps promote sound decision-making