**********************************************************************
**** Call for Papers and Participation ****
**** Workshop on Applications of Multiple-Instance Learning ****
**** At the 23rd International Conference on Machine Learning ****
**** Carnegie Mellon University ****
**** June 29, 2006 ****
**********************************************************************
http://www.cs.wisc.edu/~sray/miws.html
The Workshop on Applications of Multiple-Instance Learning, to be held
at ICML 2006, will focus on novel methods and innovative applications
of multiple-instance (MI) learning. In this setting, concepts are
learned from labeled bags (multisets) of instances rather than single
instances. The labels of the individual instances are unknown. However,
there are constraints that relate instance labels to the labels of
the bags.
This workshop will bring together leading researchers in MI learning and
researchers in potential application domains for MI learning. The goals of
this workshop are: (i) to explore novel methods and application domains
of MI learning, (ii) to explore challenging research issues that arise
when applying MI algorithms in practice, and (iii) to plan the creation
of a problem repository for MI problems for researchers to work with. To
accomplish the second goal, we are inviting leading researchers from
several domains where MI has been applied, including text, computer vision
and bioinformatics, to talk about their work and discuss connections to
MI learning and the challenges involved. To accomplish the first goal,
submissions are invited to the workshop describing novel and innovative
work in MI methods and applications. Submissions should be formatted
using the ICML conference format and emailed to the contact person
(sray_AT_eecs.oregonstate.edu) no later than April 28, 2006. Additional
information, news and updates about the workshop is be available at
http://www.cs.wisc.edu/~sray/miws.html .
Workshop format (subject to change)
1. Invited Talk on state-of-the-art in MI learning
2. Invited Talks on MI application domains (Prof. Mark Craven, University
of Wisconsin, Madison; Prof. Amy McGovern, University of Oklahoma;
Prof. Ziv Bar-Joseph, CMU)
3. Technical Paper Session
4. Panel Discussion on methods and domains for MI learning
Program Committee
Peter Auer, University of Leoben
Sally Goldman, Washington University in St. Louis
Soumya Ray, Oregon State University
Stephen Scott, University of Nebraska
Important Dates:
Paper submission dealine: April 28, 2006
Notification of acceptance: June 2, 2006
Camera-ready copies: June 9, 2006
Workshop date: June 29, 2006
|