[Colloquium] TODAY 1-2 pm, Dan McDonald, Machine Learning Seminar

[Annie Simmons]

University of Chicago and Toyota Technological Institute at Chicago
Machine Learning Seminar Series


Dan McDonald
UChicago Statistics

Wednesday, May 29, 2019 at 1:00 - 2:00 pm
Saieh Hall of Economics (SHFE) Room 203


Title:
Trend Filtering in exponential families 

Abstract:
Trend filtering is a modern approach to nonparametric regression thatis more adaptive to local smoothness than splines or basisprocedures. Current analysis of trend filtering focuses on estimatinga function corrupted by Gaussian noise, but our work extends thistechnique to general exponential family distributions. This extensionis motivated by the need to study massive, gridded climate dataderived from polar-orbiting satellites. We present algorithms tailoredto large problems, theoretical results for general loss functions, andprincipled methods for tuning parameter selection without excesscomputation

Bio:
Daniel McDonald, Visiting Assistant Professor of Statistics, studies applications of statistical machine learning. Most of his work involves providing theoretical justification for existing methodology. He is also interested in computational approximations; time series; and applications in economics, climate science, and chemistry. McDonald is visiting from Indiana University, Bloomington where he is Associate Professor of Statistics and Adjunct Associate Professor of Computer Science.

Host: Rebecca Willett







Annie Simmons
Project Assistant IV
Computer Science Department
John Crerar Library Building
5730 S. Ellis
Chicago, IL 60637 
773.834.2750
773.702.8487
simmons3 at cs.uchicago.edu

“The dream is free the hustle is sold separately"

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20190529/4edb7df0/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Dan McDonald.pdf
Type: application/pdf
Size: 899023 bytes
Desc: not available
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20190529/4edb7df0/attachment-0001.pdf>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20190529/4edb7df0/attachment-0003.html>