A0183
Title: On the minimax rate of the Gaussian sequence model under (bounded) convex constraints
Authors: Matey Neykov - Northwestern University (United States) [presenting]
Abstract: The purpose is to discuss the exact minimax rate of a Gaussian sequence model under (bounded) convex constraints, purely in terms of the local geometry of the given constraint set. It is argued that the local entropy of the set K determines the rate via a certain fixed point equation. The results are extendable to nonparametric density estimation as well as nonparametric regression.
