[Colloquium] TTIC Talks: Garvesh Raskutti, UC Berkeley

[Liv Leader]

*REMINDER:*

When:     Wednesday, February 1 @ 11 a.m.

Where:   TTIC Conference Room #526, 6045 S. Kenwood Avenue, 5th Floor

Who:      Garvesh Raskutti, UC Berkeley

Title:       Inference for large-scale models over kernel classes:
Computational and statistical aspects

Abstract
Large-scale inference problems, where the number of
features is growing with the sample size, have recently received a
significant amount of attention and research. Most of the past research on
the theory and methodology for large-scale inference problems has
involved assuming the data follows a parametric model. However for many
applications, the assumption that the response has a known parametric form
may be too restrictive. Kernel-based approaches provide an efficient way
to extend algorithms for parametric inference to algorithms for
non-parametric inference. My talk consists of two parts, both addressing
computational and statistical aspects of large-scale inference problems
using kernel methods. Firstly, I provide an analysis of sparse additive
models over kernel classes, a non-parametric generalization of sparse
linear models. In particular, I present a polynomial-time algorithm for
estimating sparse additive models that achieves minimax optimal rates in
mean-squared error. The second problem provides analysis of the early
stopping strategy applied to non-parametric regression over kernel
classes. My analysis yields a computable data-dependent stopping rule that
has minimax optimal mean-squared error rate. Both projects I present are
based on joint work with Professors Martin Wainwright and Bin Yu.

Host: Nati Srebro: nati at ttic.edu

-- 
Liv Leader
Human Resources Coordinator

Toyota Technological Institute Chicago
6045 S Kenwood Ave
Chicago, IL 60637
Phone- (773) 702-5033
Fax-     (773) 834-9881
Email-  lleader at ttic.edu <jam at ttic.edu>
Web-   www.ttic.edu
<http://www.ttic.edu/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20120131/e27a4941/attachment.htm