The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Abstract Details
|
Activity Number:
|
219
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 1, 2011 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #300109 |
|
Title:
|
Estimation of High-Dimensional Low-Rank Matrices
|
|
Author(s):
|
Alexandre Tsybakov*+
|
|
Companies:
|
Laboratoire de Statistique, CREST
|
|
Address:
|
, , ,
|
|
Keywords:
|
high-dimensional statistics ;
sparsity ;
matrix completion ;
optimal rates of convergence ;
low rank matrix estimation
|
|
Abstract:
|
This talk considers the model of trace regression, in which one observes linear combinations of entries of an unknown matrix corrupted by noise. We are particularly interested in high-dimensional setting where the dimension of the matrix can be much larger than the sample size. This talk discusses the estimation of the underlying matrix under the assumption that it has low rank, with a particular emphasis on noisy matrix completion. We consider several estimators, we derive non-asymptotic upper bounds for their prediction and estimation risks, and we show their optimality in a minimax sense on different subclasses of matrices satisfying the low rank assumption.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2011 program
|
2011 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.