ATG seminar series presents
Some Interesting Problems in Social Computation
by Michael Rovatsos from the University of Edinburgh
Time: Wednesday, Oct. 8 at 14:30 in 205.
In this talk I'll be reporting on our recent work in the SmartSociety project, where we try to develop techniques for building large-scale collective adaptive systems composed of humans and artificial agents. After presenting an initial architecture and prototypical implementation of the kinds of systems we're envisioning, I'll discuss several interesting problems related to algorithmic, architectural, and conceptual modelling issues. Some of these will be more concrete and well-defined and relate to research in planning, recommender systems, and agent interaction protocols, but some involve more blue-sky, speculative ideas extending to issues related to security and privacy, collective intelligence, and even the prospect of "evil" (social) AI.
Michael Rovatsos is a senior lecturer at School of Informatics at The University of Edinburgh, and Director of the Centre For Intelligent Systems and their Applications.