Overview‎ > ‎

Call for papers

Confluence between Kernel Methods and Graphical Models 
workshop held in conjunction with Neural Information Processing Systems (NIPS 2012)

December 7 or 8 (TBD), 2011, Lake Tahoe, Nevada, USA.

This workshop seeks to bring together leading researchers in machine learning interested in modeling complex data by combining kernel and graphical model techniques. The workshop will be a mix of invited talks, selected research presentations, posters and a panel discussion, with plenty of time to discuss open questions, new ideas, explore potential collaborations,
etc. The workshop is part of the Neural Information Processing Systems
conference (http://nips.cc/).

This workshop addresses three main research questions: first, how may kernel methods be used to address difficult learning problems for graphical models, such as inference for multi-modal continuous distributions on many variables, and dealing with non-conjugate priors? Second, how might kernel methods be advanced by bringing in concepts from graphical models, for instance by incorporating sophisticated conditional independence structures, latent variables, and prior information? Third, how to make kernels combined with graphical models scale to deal with big data?

This workshop aims to "connect the dots" and explore a unified framework to address a broad range of learning problems, to the mutual benefit of reseachers in kernels and graphical models. The goals of the workshop are: 
* to provide an accessible review and synthesis of recent results combining graphical models and kernels, and
* to provide a discussion forum for open problems, technical challenges, and applications in new and emerging areas. 

We invite submissions which address learning problems using both graphical models and kernel methods. Potential topics include, but are not limited to,
the following areas:
1) Gaussian processes
2) Nonparametric Bayes methods 
3) Sampling techniques with kernels
4) Hilbert space embedding of distribution and conditional distribution
5) Nonparametric graphical models
6) Scalable methods for big data


We encourage attendees for contribution of an extended abstract (up to
4 pages using NIPS style). Abstracts should be sent by email to
kernelgraphical.nips2012@gmail.com, not later than September 26, 2012.
Organizers will review and select submissions by Oct 14, 2012. Accepted
submissions will be presented as a talk or posters.