Competition and Cooperation in Multi-Agent Systems

Professors:Kevin Leyton-Brown (University of British Columbia), Vincenzo Bonifaci (IASI - CNR), Alessandro Farinelli (University of Verona)
Credits:2.5
Range:B
Note:

Full description

Competition and Cooperation in Multi-Agent Systems  

 

Prof. Kevin Leyton-Brown (University of British Columbia) 

Dr. Vincenzo Bonifaci (IASI - CNR) 

Dr. Alessandro Farinelli (University of Verona)

 

Competitive multiagent systems are environments in which two or more agents interact, and agents' desires cannot be relied upon to coincide. Lately, computers have started to form competitive multiagent systems, too: large-scale distributed systems like the internet or autonomous robotic systems are becoming ubiquitous,  and in such systems users cannot simply be trusted to behave responsibly or they simply have competitive goals. In this set of seminars we will present game theoretical formalisms and design principles of multiagent protocols that achieve desirable properties. Special emphasis will be given to resource allocation problems  that can be modeled as computerized auctions. Furthermore, we'll address the problems of answering these game theoretical questions computationally and to finding compact representations of such games.  The problems of designing competitive multiagent robotic systems and  competitive mechanisms with suitable game theoretical properties that optimize specific objectives will also be addressed. 

---- 

Detailed program:

Monday, Oct. 29 -  Dr. Vincenzo Bonifaci: Models of Selfish Agents

14:00 - 16:30, Room A4

Wednsday, Oct.31 -  Dr. Vincenzo Bonifaci: Design of Mechanisms for Agent Cooperation

14:00 - 16:30, Room A4

Monday, Nov. 5th -  Prof. Kevin Leyton-Brown:  Beyond Equilibrium: Predicting Human Behavior in Normal Form Games

16:00 - 18:30, Aula Magna

Wednesday, Nov 7th -  Prof. Kevin Leyton-Brown: Computational Mechanism Analysis: Towards a "CPLEX for Mechanisms"

16:00 - 18:30, Aula Magna

Thursday, Nov 8th - Prof. Kevin Leyton-Brown: Scaling Up Game Theory: Representation and Reasoning with Action Graph Games

16:00 - 18:30, Room B203

Friday, Nov 9th - Dr. Alessandro Farinelli: Optimization in Multi-Agent Systems based on DCOP

14:00 - 16:30, Room A4

Monday, Nov 12th - Dr. Alessandro Farinelli: Optimization in Multi-Agent Systems based on DCOP

14:00 - 16:30, Room A4

-------------------------

 

Short Bio of Kevin Leyton-Brown is an associate professor in computer science at the University of British Columbia. He holds a PhD and M.Sc. from Stanford University (2003; 2001) and a B.Sc. from McMaster University (1998). Much of his work is at the intersection of computer science and microeconomics, addressing computational problems in economic contexts and incentive issues in multiagent systems. He also studies the application of machine learning to the automated design and analysis of algorithms for solving hard computational problems. He has co-written two books, "Multiagent Systems" and "Essentials of Game Theory," and eighty peer-refereed technical articles. With his coauthors, he has received best paper awards from JAIR, ACM-EC and LION, and numerous medals for the portfolio-based SAT solver SATzilla at the International SAT Competition. He is program chair for the ACM Conference on Electronic Commerce (ACM-EC), and associate editor of the Journal of Artificial Intelligence Research (JAIR), the Artificial Intelligence Journal (AIJ), and ACM Transactions on Economics and Computation. He split his 2010-11 sabbatical between Makerere University in Kampala, Uganda, and the Institute for Advanced Studies at Hebrew University in Jerusalem, Israel. He has served as a consultant for Trading Dynamics Inc., Ariba Inc., and Cariocas Inc., and was scientific advisor to Zite Inc until it was acquired by CNN in 2011.

  

Contacts:  

Prof. Luca Iocchi

Prof. Stefano Leonardi