Cooperation and Competition in Network Tasks

Add to your list(s) Download to your calendar using vCal

• Yoram Bachrach, Microsoft Research Cambridge, UK
• Tuesday 18 October 2011, 12:00-13:00
• Dennis Gabor Seminar Room (611), EEE Department.

If you have a question about this talk, please contact Henrik J Jensen.

Cooperative game theory is all about how selfish agents might agree to collaborate and then share their spoils. It allows answering questions such as: - Would the political power balance change if a big party decided to split into two smaller parties? - Can selfish behavior jeopardize making mutually beneficial agreements? - How might pirates share a hidden treasure when they need each other to find it?

Tools from cooperative game theory can allow us to measure the importance or power of an agent in achieving a task even in domains where only coalitions of agents can achieve the task. I will demonstrate how such “power indices” can be used to analyse decision making bodies and political parties or perform manipulations in political settings.

I will then discuss how such measures can be applied to network settings. In such domains, agents control nodes or edges in a network and attempt to achieve various goals, such as establishing communication between certain nodes or achieving a required bandwidth between two endpoints. Power indices can identify problematic areas in the network, such as reliability hotspots or unsecure areas.

This talk is part of the Complexity & Networks Group series.

This talk is included in these lists:

• Complexity & Networks Group
• Dennis Gabor Seminar Room (611), EEE Department

Note that ex-directory lists are not shown.