Friday, October 23, 2015
Theory of Computing Seminar
The Speed of Social Learning
Omer Tamuz, Caltech
We consider Bayesian agents who learn from private signals, as well as the actions of the other. We show that learning from actions is slower than learning from signals, and that that increased interaction between the agents can lower the speed of learning.
Joint with Matan Harel, Elchanan Mossel and Philipp Strack