• Description
• Important Dates
• Topics of Interest
• Submission

Description

Multiple Kernel Learning (MKL) has received significant interest in the machine learning community. It is reaching a point where efficient systems can be applied out of the box to various application domains, and several methods have been proposed to go beyond canonical convex combinations. Concurrently, research in the area of metric learning has also progressed significantly, and researchers are applying them to various problems in supervised and unsupervised learning. A common theme is that one can use data to infer similarities between objects while simultaneously solving the machine learning task. A special topic of the Journal of Machine Learning Research will be devoted to kernel and metric learning with a special emphasis on new directions and connections between the various related areas; like learning the kernel, learning metrics, and learning the covariance function of a Gaussian process. We invite researchers to submit novel and interesting contributions to this special issue.

Important Dates

Topics of Interest

Topics of interest include:

• New approaches to MKL, in particular, kernel parameterizations different than convex combinations and new objective functions
• New connections between kernel, metric and covariance learning, e.g., from the perspectives of Gaussian processes, learning with similarity functions, etc.
• Sparse vs. non-sparse regularization in similarity learning
• Efficient algorithms for metric learning
• Use of MKL in unsupervised, semi-supervised, multi-task, and transfer learning
• MKL with structured input/output
• Innovative applications

Submission Procedure

Authors are kindly invited to follow the standard JMLR format and submission procedure. The number of pages is limited to 30. Please include a note stating that your submission is for the special topic on Multiple Kernel Learning.