Scaling AI Through Multi-Agent Organizations
published: July 22, 2009, recorded: July 2009, views: 6263
Report a problem or upload filesIf you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.
Scaling remains one of the grand challenges for AI. Lesser has been using organizational control to build multiagent systems with hundreds to thousands of intelligent agents. This approach can also be used to structure complex AI systems with extensive and heterogeneous knowledge. Organizational control is a multi-level approach in which organizational goals, roles, and responsibilities are dynamically developed, distributed, and maintained to serve as guidelines for making detailed operational control decisions by the individual agents. Lesser will illustrate the use of organizational control in three distributed application areas: (1) an adaptive sensor and interpretation vehicle-tracking network, (2) a peer-to-peer information search and retrieval system, and (3) a self-improving task allocation system. He will highlight the important balance between externally-directed and self-directed agent activities in uncertain and dynamic environments. Then he will present the continuing research challenges, including how to automate the design of an organization and evolve it as conditions change, create an organizationally situated agent, and evaluate and predict an organization’s performance.
Download slides: ijcai09_lesser_saitmao_01.pdf (24.8 MB)
Download slides: ijcai09_lesser_saitmao_01.ppt (4.4 MB)
Link this pageWould you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !