Associate Professor in Electronics and Computer Science
University of Southampton
The first few hours in the aftermath of a disaster are critical for emergency responders to save as many lives as possible. Increasingly, responders rely on crowdsourced reports, unmanned aerial vehicles, and mobile coordination systems to carry out their tasks in very dynamic, uncertain, and dangerous conditions. The challenge for researchers is to build technology that is supports the work of emergency responders and not hinder it. In my talk, I’ll present the work I’ve led in the last 10 years on agent-based technologies for disaster response, with a focus on how our methodology has evolved to encompass studies of such technologies ‘in the wild’.
Dr. Sarvapali Ramchurn is an Associate Professor in the Agents, Interaction, and Complexity research group where he carries out research into the design of autonomous agents and multi-agents for real-world socio-technical applications including energy systems, disaster management, and crowdsourcing. He works closely with industry and his research touches on a number of fields including Machine Learning, Data Science, and Game Theory. Specifically, he has pioneered the development of agent-based coordination algorithms for distributed task allocation that have been deployed on real-world unmanned aerial vehicles and in the Premier League’s Fantasy Football game where his approach has been shown to outperform more than 2M human players. His papers have been cited more than 2000 times (according to Google scholar) and his work has featured in various media including BBC News, New Scientist, EPSRC Pioneer, and Wired.
Tis work explores a PAC (probably approximately correct) learning model in cooperative games. Specifically, we are given m random samples of coalitions and their values, taken from some unknown cooperative game; can we predict the values of unseen coalitions? We study the PAC learnability of several well-known classes of cooperative games, such as network flow games, threshold task games, and induced subgraph games. We also establish a novel connection between PAC learnability and core stability: given an unknown cooperative game, it is possible to find payoff divisions that are likely to be stable using a polynomial number of samples.
This is joint work with Maria F. Balcan and Ariel D. Procaccia, and will be presented at IJCAI 2015 (Game Theory 3 Session, July 28th 15:10 – 16:30, room LB1).
Yair Zick is a postdoctoral research fellow in the computer science department of Carnegie Mellon University.
His research interests span cooperative game theory, computational social choice and their applications to domains such as security, privacy,
machine learning and education.
He is the recipient of the AAMAS 2011 Pragnesh Jay Modi best student paper award, and the 2014 IFAAMAS Victor Lesser distinguished dissertation award.