Annonce

Les commentaires sont clos.

Object, functional, structured data : towards next generation kernel-based methods

14 Mai 2012


Catégorie : Conférence internationale


Object, functional, structured data : towards next generation kernel-based methods. ICML 2012 Workshop, June 30, 2012, Edinburgh, UK.

This workshop concerns analysis and prediction of complex data such as objects, functions and structures. It aims to discuss various ways to extend machine learning and statistical inference to these data and especially to complex outputs prediction. A special attention will be paid to operator-valued kernels and tools for prediction in infinite dimensional space.

 

Important dates

  • Submission due by May 16, 2012.
  • Author Notification, May 21, 2012.
  • Workshop, June 30, 2012.

Topic

This workshop concerns analysis and prediction of complex data such as objects, functions and structures. It aims to discuss various ways to extend machine learning and statistical inference to these data and especially to complex outputs prediction. A special attention will be paid to operator-valued kernels and tools for prediction in infinite dimensional space.

Context and motivation

Complex data occur in many fields such as bioinformatics, information retrieval, speech recognition, image reconstruction, econometrics, biomedical engineering. In this workshop, we will consider two kinds of data: functional data and object or structured data. Functional data refers to data collected under the form of sampled curves or surfaces (longitudinal studies, time series, images). Analysis of these data as samples of random functions rather that a collection of individual observations is called Functional Data Analysis (FDA). FDA involves statistics in infinite-dimensional spaces and is closely associated to operatorial statistics. Its main approaches include functional principal component analysis and functional regression. Many theoretical challenges remain open in FDA and attract an increasing number of researchers.

Object and structure data exhibit an explicit structure like trees, graphs or sequences. For instance, documents, molecules, social networks and again images can be easily encoded as objet structured data. For the two last decades, both machine learning and statistics communities have developed various approaches such as graphical probabilistic models as well as kernel methods to take into account the structure of the data. In the meantime, FDA has been extended to Object Data Analysis which deals with samples of object data.

However, most of the efforts have been concentrated so far on dealing with complex inputs. In this workshop, we would like to emphasize the problem of complex outputs prediction which is involved for instance in multi-task learning, structured classification and regression, and network inference. All these tasks share a common feature: they can be viewed as approximation of vector-valued functions instead of scalar-valued functions and in the most general case, the output space is an Hilbert space. A promising direction first developed in (Micchelli and Pontil, 2005) consists in working with Reproducing Kernel Hilbert Spaces with operator-valued kernels in order to get an appropriate framework for regularization. There is thus a strong link between recent works in machine learning about prediction of multiple or complex outputs and functional and operatorial statistics.

This workshop aims at bringing together researchers from both communities to  1) provide an overview of existing concepts and methods, 2) identify theoretical challenges and (3) discuss practical applications and new tasks. 

Invited speakers

  • Yasemin Altun (Google)
  • Frédéric Ferraty (University of Toulouse, France)
  • Arthur Gretton (Gatsby Unit, UCL MPI for Intelligent Systems, UK)
  • Neil Lawrence (University of Sheffield, UK)
  • Steve Marron (University of North Carolina, USA)
  • Charles Micchelli (University of Albany, USA)

Call for contributions

We invite short, high-quality submissions on the following topics: 

  • complex output learning
  • structured output prediction
  • functional data analysis
  • object data analysis
  • operator-valued kernels
  • operator-based statistics
  • joint-kernel maps
  • statistical dynamics 
  • applications (non exhaustive list) : signal and image processing, bioinformatics, natural language processing, time series modeling …

Submission guidelines

Submissions should be written as extended abstracts, no longer than 4 pages in the ICML latex style.  ICML style files and formatting instructions can be found at  The submissions should include the authors' name and affiliation since the review process will not be double blind. The extended abstract may be accompanied by an unlimited appendix and other supplementary material, with the understanding that anything beyond 4 pages may be ignored by the program committee. Please send your submission by email to nextgenkernelicml2012@gmail.com  before May 7, 2102 at midnight PDT.   Recently-published work is allowed.

We expect to select contributions for the spotlight and poster sessions. Authors will receive a notification by May 21, 2012. 

Organizers

  • Florence d’Alché-Buc (University of Evry & INRIA-Saclay, France)
  • Hachem Kadri (INRIA-Lille, France)
  • Massimiliano Pontil (University College London, UK)
  • Alain Rakotomamonjy (University of Rouen, France)

https://sites.google.com/site/nextgenkernels/

Website admin: Céline Brouard (University of Evry, France)

Contact: nextgenkernelicml2012@gmail.com