From the archives
Empirical Likelihood for Optimistic Algorithms in Dynamic Resource Allocation
Historical material. Affiliations, schedules and roles on this page describe their original context.
Context: Paris X Nanterre
Resume:
Bandit models, and especially the UCB algorithms, are presented together with statistical challenges they involve: non-asymptotic estimation, self-normalized deviations, Empirical Likelihood.
Slides: kickoffSPADRO.pdf Date: April, 2014 Event url: ANR SPADRO kick-off meeting Keywords: