Optimizing Collective Behavior in Cultural Markets

Friday, August 4, 2017, 11:00 am - 12:00 pm PSTiCal
11th floor large conference room
This event is open to the public.
AI Seminar - Interview talk
Andres Abeluk

Every day people make a staggering number of decisions about what to buy, what to read and where to eat. The interplay between individual choices and collective opinion is responsible for much of the observed complexity of social behaviors. The impact of social influence on the behavior of individuals may distort the quality perceived by the customers, making quality and popularity out of sync. Fewer cultural products are consumed overall and some high-quality products may go unnoticed, while lower-quality ones are oversold. Understanding how people respond to this information will enable us to predict social behavior and even steer it towards desired goals.

In this talk, we take that step forward by studying how and to what extent one can optimize a cultural market in order to reduce the unpredictability and improve the efficiency of the market. Our hypothesis provides an interesting contrast with the conventional wisdom: It is not social influence per se that makes markets unpredictable, rather than the way it is used. We introduced a generative model and study the dynamics of cultural markets for different design choices on how items are displayed to the participants. Furthermore, we put to experimental test our theoretical results and show a policy that mitigates the disparities between popularity and quality that emerge from social and position biases.



Andrés Abeliuk is a postdoctoral associate at MIT Media Lab, in the Scalable Cooperation group. His research interests are in computational social science, focusing on collective intelligence. By combining an array of tools including modeling, optimization, game theory, and online experiments, Andrés attempts to explain the emergence of collective behavior. Before that, he worked as a Ph.D. student researcher in the Optimisation group at DATA61 (formerly known as NICTA), designing novel algorithms and mechanisms to harness collective behavior towards more efficient social outcomes. He holds a Ph.D. in Computer Science from The University of Melbourne (Australia).



« Return to Upcoming Events