• Description
• Schedule

News

News 12/15/08: Abstracts Posted

The submitted abstracts for the accepted talks and posters have been linked to in the schedule below. Once we have collected all the presented slides, they will also be posted.

News 09/25/08: Call for Submissions

Submissions are solicited for a Kernel Learning workshop to be held on December 13th, 2008 at this year's NIPS workshop session in Whistler, Canada. Submissions to the workshop should be on the topic of automatic kernel selection or more broadly feature selection, multi-task learning and multi-view learning.

Submissions should concern using sampled data to select or learn a kernel function or kernel matrix appropriate for the specific task at hand. Submissions are not constrained regarding the area in which which the learned kernel is applied: this can include including classification, regression, and ranking, where the use of kernels is ubiquitous, and different settings including inductive, transductive, or semi-supervised learning. Presentations on the closely related topics of feature selection, multi-task learning and multi-view learning are also encouraged. Submissions should focus on theoretical, algorithmic or large-scale empirical results.

Accepted submissions will result in a 25 minute talk or a poster presentation. The deadline for submission is October 24th and notifications will be sent out by November 7th. Submissions should be sent in NIPS format (maximum 4 pages) to rostami (at) cs (dot) nyu (dot) edu. The program committee will consist of the workshop organizers and the invited speakers

News 09/12/08: Workshop Date

The workshop has been scheduled for December 13th, and will be located in Whistler, B.C. at the Hilton Whistler and Spa. This is part of a two day workshop (December 12-13), held after the main NIPS conference.

Workshop Description

Kernel methods are widely used to address a variety of learning tasks including classification, regression, ranking, clustering, and dimensionality reduction. The appropriate choice of a kernel is often left to the user. But, poor selections may lead to sub-optimal performance. Furthermore, searching for an appropriate kernel manually may be a time-consuming and imperfect art. Instead, the kernel selection process can be included as part of the overall learning problem. In this way, better performance guarantees can be given and the kernel selection process can be made automatic.

In this workshop, we will be concerned with using sampled data to select or learn a kernel function or kernel matrix appropriate for the specific task at hand. We will discuss several scenarios, including classification, regression, and ranking, where the use of kernels is ubiquitous, and different settings including inductive, transductive, or semi-supervised learning.

The goal is to cover all questions related to the problem of learning kernels: different problem formulations, the computational efficiency and accuracy of the algorithms that address these problems and their different strengths and weaknesses, and the theoretical guarantees provided. What is the computational complexity? Does it work in practice? Equally as important, we will discuss improvements, new algorithms and motivate further interesting scenarios. The formulation of some other learning problems, e.g. multi-task learning problems, is often very similar. These problems and their solutions will also be discussed in this workshop.

Schedule

PDF w/ abstracts

Poster Presentations:

-Ravi S. Ganti, Nikolaos Vasiloglou and Alexander Gray: Hyperkernel Based Density Estimation
-Andrew G. Howard and Tony Jebara: Learning Large Margin Mappings
-S. Mosci, M. Santoro, A. Verri, S. Villa and L. Rosasco: A New Algorithm to Learn an Optimal Kernel Based on Fenchel Duality
-Hua Ouyang and Alexander Gray: Learning Nearest-Neighbor Classifiers with Hyperkernels
-Nikolaos Vasiloglou, Alexander G. Gray and David V. Anderson: Learning Isometric Separation Maps
All other submitted talks are also encouraged to give posters.