DM&ML Kickoff Workshop

Data Mining and Machine Learning
Tutorials
Kickoff Workshop
Initial Working Group Meetings
September 6-10, 2003

General Information

The Kickoff Workshop for the SAMSI Program on Data Mining and Machine Learning (DM&ML) will be held on Sunday-Tuesday, September 7-9, 2003, at the Radisson Governors Inn in Research Triangle Park, NC. It will be preceded, on Saturday, September 6, by tutorials and followed, on Wednesday, September 10, by initial meetings of the working groups that will form the core of the program.

The goal of the Kickoff Workshop is to frame issues to be addressed during the program, and the direct tie with the September 10 Working Group meetings is meant to maximize the effects of the workshop on the scientific structure of the program.

In order to stimulate engagement of all participants, the DM&ML Kickoff Workshop will have only 10 invited talks, which will be complemented by a variety of other sessions that give all attendees a chance to participate.

The tutorials will be held at the MCNC-RDI Auditorium in Research Triangle Park, NC. The workshop will be held at the Radisson Governors Inn in Research Triangle Park. The working group meetings will be held at SAMSI.

Program

PLEASE NOTE: The tutorials will be held at the MCNC-RDI Auditorium!!

Saturday, September 6: Tutorials

1:00-2:30 PM	Large p, Small n Inference David Banks, Duke University
3:00-4:30	Support Vector Machines J.S. Marron, SAMSI and University of North Carolina at Chapel Hill

PLEASE NOTE: The tutorials will be held at the MCNC-RDI Auditorium!!

Sunday, September 7

9:30 AM	Welcome and Introductions
10:00	Similarities and Differences between Statistics, Machine Learning and Data Mining Leo Breiman, University of California, Berkeley
11:00	Convex Optimization and Variational Inference Algorithms: Alternatives to MCMC for Large-scale Statistical Models Micheal I. Jordan, University of California, Berkeley
12:00 N	Lunch
1:00-3:00 PM	Birds-of-a-Feather Sessions (Precursors of Working Groups) These will reflect workshop and participant interests. Current candidates are "Large p, small n inference," Bioinformatics, Support Vector Machines, Computational Experiments, Text Mining and Model Selection
3:30	Statistical Methods for Text Mining David Madigan, Rutgers University
4:30	Using Proc MULTEST of SAS/STAT for Data Mining Peter Westfall, Texas Tech University

Monday, September 8

9:30 AM	Temporal Data Mining: Novel Algorithms and Their Applications K. P. Unnikrishnan, General Motors Research and Development
11:00	Postmarketing Drug Adverse Event Surveillance and the Innocent Bystander Effect William DuMouchel, AT&T Labs Research
12:00 N	Lunch
1:00-2:00 PM	Poster Sales Talks (2 minutes each)
2:30-3:30	Data Mining in Anti-Terrorism Applications Jeff Schneider, Carnegie Mellon University
4:00-5:30	Second Chance Seminar (Anyone can talk)
7:00-9:00	Reception & Poster Session at SAMSI

Tuesday, September 9

9:30 AM	Statistical Tools for the Sciences Leo Breiman, University of California, Berkeley
11:00	Using Graphics in Exploratory Data Analysis and Data Mining: An Application of Supervised Classification in Olive Oil Quality Di Cook, Iowa State University
12:00 N	Lunch
1:00-3:00 PM	Young Researchers Session
3:30-4:30	Bayesian Additive Regression Trees Robert McCulloch, University of Chicago
4:30-5:00	Final Discussion
5:00 PM	Adjourn

Wednesday, September 10

10:00 AM-4:00 PM

Working Group Meetings at SAMSI

Scientific Committee

David Banks (Duke; Co-Chair), Mary Ellen Bock (Purdue), Jerome Friedman (Stanford), Alan F. Karr (NISS; Chair), David Madigan (Rutgers), William DuMouchel (AT&T), Warren Sarle (SAS Institute)

Local Information (Housing, ...)

DM&ML Program Home Page

SAMSI Home Page