[Colloquium] TTIC Colloquium: Peter Grunwald, CWI Amsterdam

[Liv Leader]

When:       *Tomorrow*, Thursday, June 21 @ 11 a.m.

Where:      TTIC, 6045 S. Kenwood Avenue, Room 526

Who:         Peter Grunwald, CWI Amsterdam, Mathematical Institute, Leiden
University

Title:          Safe Learning: How to Modify Bayesian Inference when All
Models are Wrong

Standard Bayesian inference can behave badly if the model under
consideration is wrong:  in some simple settings, the posterior fails to
concentrate even in the limit of infinite sample size.  We introduce a test
that can tell from the data whether we are in such a situation. If we are,
we can adjust the learning rate in a data-dependent way. The resulting
"safe" estimator continues to achieve good rates with wrong models. We
provide PAC-Bayesian style theorems showing that, e,g,, when applied to a
union of classification models, the safe estimator converges at the optimal
rate for the Tsybakov exponent of the underlying distribution. This
establishes a connection between Bayesian inference and statistical
learning theory.

Host: David McAllester, mcallester 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/20120620/8e485b1b/attachment.htm