Conference abstracts

Session B4 - Learning Theory

July 13, 18:00 ~ 18:25 - Room B6

Optimal graphon estimation

Olga Klopp

University Paris Nanterre, France   -   kloppolga@math.cnrs.fr

In the last decade, network analysis has become an important research field driven by applications. We study the twin problems of estimating the connection probability matrix of an inhomogeneous random graph and the graphon of a $W$-random graph. We consider classes of block constant matrices and step function graphons and establish the minimax estimation rates with respect to different distances: $l_1$, $l_2$ and the cut metric. We also extend our study to the case of unbounded graphons.

Joint work with Nicolas Verzelen (INRA, France).

View abstract PDF



FoCM 2017, based on a nodethirtythree design.