[UAI] selecting and combining models workshop / Feb.1st deadline

From: Yoshua Bengio (bengioy@IRO.UMontreal.CA)
Date: Mon Jan 24 2000 - 14:16:05 PST

  • Next message: frey@dendrite.uwaterloo.ca: "Re: [UAI] sigmoid belief nets"

            ml@ics.uci.edu
    Cc: dale@cs.uwaterloo.ca
    Mime-Version: 1.0
    Content-Type: text/plain; charset=iso-8859-1
    Content-Transfer-Encoding: 8bit
    X-Mailer: Mutt 0.69

    Hello,

    This is a reminder that the deadline for submitting abstracts to
    the workshop on model selection and combination (Montreal, April 12-14 2000)
    is in one week, Monday February 1st, 2000. The submission is
    electronic (via e-mail). Registration to the workshop will be free.

    --------------------------------------------------------------------

                 Call for Presentations: CRM Workshop on

     Selecting and Combining Models with Machine Learning Algorithms

                              April 12-14, 2000
                     Centre de Recherches Mathematiques
                                 Montreal

    Organizers: Yoshua Bengio (Universite de Montreal) and
                Dale Schuurmans (University of Waterloo)

    A central objective of machine learning research is to develop
    algorithms that learn predictive relationships from data. This is a
    central component of data mining and knowledge discovery tasks, which
    are becoming commonplace applications in the realm of e-commerce.
    This is a difficult task, however, because inferring a predictive
    function from data is in fact an "ill-posed" problem; that is, many
    functions can often "fit" a given finite data set, and yet these
    functions might generalize very differently on new data drawn from the
    same distribution.

    To make this problem well-posed one needs to somehow "calibrate" the
    complexity of the proposed function class to the amount and quality of
    available sample data. A classical approach is to perform "model
    selection" where one imposes a preference structure over function
    classes and then optimizes a combined objective of class preference
    and data fit. In doing so, however, it would be useful to have an
    accurate estimate of the expected generalization performance at each
    preference level; one could then pick the function class that obtained
    the lowest expected error, or combine functions from the functions
    classes with the lowest expected error, and so on. Many approaches
    have been proposed in the past for this purpose, both in the
    statistics and the machine learning research communities.

    Recently in machine learning there has been significant interest in
    new techniques for evaluating generalization error, for optimizing
    generalization error, and for combining and selecting models. This is
    exemplified, for example, by recent work on Structural Risk
    Minimization, Support Vector Machines, various Boosting algorithms,
    and the Bagging algorithm. These new approaches suggest that better
    generalization performance can be obtained using new, broadly
    applicable procedures. Progress in this area has not only been
    important for improving our understanding of how machine learning
    algorithms generalize, it has already been demonstrated to be very
    useful for practical applications of machine learning and data
    analysis.

    This workshop will bring together several key researchers in the
    fields of machine learning and statistics to present their recent
    results and debate the controversial issues that have been dividing
    them in the recent machine learning and neural network conferences.
    The following leaders in this field have tentatively accepted to
    participate to the workshop as invited speakers:

    Peter Bartlett (Australia National University), Leo Breiman
    (University of California-Berkeley), Tom Dietterich (Oregon State
    University), Yoav Freund (AT&T Labs-Research), Radford Neal
    (University of Toronto), Michael Perrone (IBM T.J. Watson Research
    Center), Robert Schapire (AT&T Labs-Research), Grace Wahba (University
    of Wisconsin at Madison).

    The workshop will be sponsored by the CRM (Centre de Recherches
    Mathematiques) as well as by the MITACS (Mathematics of Information
    Technology And Complex Systems) Network of Centers of Excellence.
     
    Contributors to the workshop are invited to submit a short (1 or 2
    page) summary in electronic form (ascii text, postscript or pdf) of
    the proposed presentation by e-mail, to one of the organizers by
    February 1st, 2000: bengioy@iro.umontreal.ca or dale@cs.uwaterloo.ca.

    Information on the workshop will be posted on the following web site:
     www.iro.umontreal.ca/~bengioy/crmworkshop2000

    --------------------------------------------------------------------

    -- 
    Yoshua Bengio 
    Professeur aggrégé
    Département d'Informatique et Recherche Operationnelle
    Université de Montréal, 
    addresse postale: C.P. 6128 Succ. Centre-Ville, Montreal, Quebec, Canada H3C 3J7
    addresse civique: 2920 Chemin de la Tour, Montreal, Quebec, Canada H3T 1J8, #2194
    Tel: 514-343-6804. Fax: 514-343-5834. Bureau 3339.
    http://www.iro.umontreal.ca/~bengioy
    http://www.iro.umontreal.ca/~lisa
    



    This archive was generated by hypermail 2b29 : Mon Jan 24 2000 - 14:17:19 PST