Speaker:Constantinos Daskalakis

Date:12/01/2016

University:MIT

Room :A56

Time:14:00

Abstract:Game Theory and Learning have enjoyed a long history of interactions over 60+ years. In this talk, I will discuss one of their earliest interactions, providing a counter-example to Karlin's conjecture on the convergence rate of fictitious play. In the second part of the talk, I will present a novel point of interaction between the two fields, developing statistical learning algorithms for learning from strategic data sources.