Seminars & Colloquia
Sandia National Laboratories
"Agent Based Models, Social Phenomena and Massively Multiplayer
Monday November 05, 2012 11:00 AM
Location: 3211, EBII NCSU Centennial Campus
(Visitor parking instructions)
This talk is part of the
Future of Games Series
At Sandia National Labs, Agent Based Modeling (ABMs) allows for scientists to explore how political, technological, or natural events shift social behavior in ways that impact national security. Severe challenges exist in creating these models due to problems in allocating high-volumes of data from the real-world to inform and validate these models.
Massively Multiplayer Online Games (MMOGs) have recently emerged as a tractable way to acquire data on social phenomena, because these virtual environments are able to capture a great amount of data and at high- fidelity, often tracking the actions of many individuals at a time resolution of seconds. As well, the social structures expressed in MMOGs reflect those employed in the real-world, leading to mirroring behaviors that make MMOGs an attractive data source for ABM validation.
This talk will cover research taking place at Sandia National Labs in using MMOGs to improve Agent Based Modeling construction and validation. In particular, we will discuss recent research on analyzing the motivations for people to participate in a Large Scale Conflict (LSCs), and the correlates between theoretical frameworks generated from known civil wars to what we have observed within a virtual world.
Dr. Kiran Lakkaraju is a Senior Member of the Technical Staff with a background in artificial intelligence, multi-agents systems and computational social science. He holds a M.S. and Ph.D. in Computer Science from the University of Illinois. Kiran's primary research interests lie in developing computational models of behavior change in society such as the creation and propagation of linguistic behavior within a population. More recently he has focused on developing models that explore the link between social structure (social networks, roles/hierarchy) and cognitive structure (how concepts are interrelated, cognitive consistency, confirmation bias) with respect to problems of information dissemination and attitude change. A critical issue in making models useful is validation how well the model corresponds to reality. To this end, he has been looking at how to run large, online, experiments on attitude and behavior change using existing crowdsourcing tools (such as Amazon Mechanical Turk) and how to derive relevant knowledge from Massively Multiplayer Online Games.
Jon Whetzel is a Senior Member of Technical Staff with a diverse background in artificial intelligence, machine learning, and video game design. He holds an M.S. in Computer Science from Texas A&M University. Jon is a principal investigator and technical lead at Sandia on several deployed training products that automate modeling team performance using data from simulation environments. Outside of Sandia, Jon is an instructor on video game development in the Interdisciplinary Film & Digital Media (IFDM) program at the University of New Mexico. He serves as the chairperson for the Albuquerque chapter of the International Game Developers Association (IGDA) and served as reviewer for the IEEE Innovation in Games conference.
Host: Michael Young, Computer Science, NCSU
Back to Seminar Listings
Back to Colloquia Home Page