[Corpora-List] 2nd CFP: Bayesian Methods in NLP Workshop at NIPS

[Hal Daume III]

************************************************************************

                            CALL FOR PAPERS

            Bayesian Methods for Natural Language Processing

                            Workshop at the 
            Neural Information Processing Systems Conference
                              (NIPS 2005)

                  http://www.isi.edu/~hdaume/BayesNLP/

               ** Submission Deadline: 21 October 2005 **

************************************************************************
                   [ Apologies for multiple postings ]


OVERVIEW
--------

Models of natural language processing problems are often incredibly
complex, and there is never enough data to properly estimate all the
required parameters. This has lead to a strong need for learning
techniques with built-in capacity control; most classical solutions to
this problem involve largely ad-hoc smoothing techniques. The
application of Bayesian learning methods to these problems could
potentially result in more effective models, for which extensive
cross-validation is no longer required for hyperparameter tuning or
model selection.

The goals of this workshop are to bring together researchers from both
the Bayesian machine learning community and the natural language
processing community to enable cross-fertilization of techniques,
models and applications. We wish to focus on the following issues:

    * Statistical Models: Current Bayesian models for text have
      largely focused on "bag of words" style approaches, where
      conditional independence is assumed between words. This leads to
      a convenient interpretation of a document as a sequence of draws
      from multinomial distributions, but does not account for any of
      the internal structure that exists in documents and which NLP
      researchers are interested in. How can we build models that move
      beyond the bag of words assumption? What structures are useful
      for modeling? How can we model these structures efficiently? Can
      we learn these models automatically?

    * Applications-oriented Models: Many statistical models for text
      have aimed at automatically inferring implicit relationship
      between varied elements of documents in a corpus. How can we use
      such models to aid in applications? Can we develop similar
      models that are aimed at solving a real-world NLP task? For what
      NLP applications are Bayesian techniques appropriate and how can
      we develop models specific to these problems?



CALL FOR PARTICIPATION
----------------------

We invite submission of workshop papers that discuss ongoing or
completed work dealing with Bayesian techniques applied to natural
language processing problems (see below for an incomplete list of
possible topics). A workshop paper should be no more than six pages in
the standard NIPS format. Authorship should not be blind. Please
submit a paper by emailing it in Postscript or PDF format to
hdaume at isi.edu with the subject line "BNLP Submission". We anticipate
accepting four to six such papers for 15 minute presentation slots
(exact details will be worked out shortly). Please only submit an
article if at least one of the authors will be able to attend the
workshop and present the work.

We are especially interested in submissions from authors in the NLP
community who have not previously attended a NIPS conference. If you
fall into this category, please note this in your email when you
submit your paper.

Relevant Topics:

    * Models that move beyond the bag-of-words assumption
    * Techniques that apply to problems other than language modeling
    * Structure-learning techniques for language
    * Bayesian extensions to well-known NLP models
    * Application of Bayesian techniques to NLP problems
    * Both supervised and unsupervised techniques are welcome

We also welcome position papers of at most two pages in length that
discuss, with appropriate argumentation, whether or not Bayesian
techniques are applicable to NLP problems and, if so, which
ones. These should be submitted in the same way as standard workshop
papers. These will be used to help guide discussion during panel
sessions.


IMPORTANT DATES
---------------

    18 Aug 05 -- Call for participation
    21 Oct 05 -- Paper submission deadline
     4 Nov 05 -- Notification of paper acceptance
    25 Nov 05 -- Survey and position paper deadlines
  9/10 Dec 05 -- Workshop in Whistler


INVITED SPEAKERS AND PANELISTS
------------------------------

Kenneth Church (Microsoft Research) -- Invited Speaker & Panelist
Tom Griffiths (Brown University)    -- Invited Speaker & Panelist
Jeff Bilmes (U. of Washington)      -- Panelist
Andrew McCallum (UMass Amherst)     -- Panelist

RESEARCHER SURVEY
-----------------

Regardless of whether you submit a paper or not, if you are a
researcher in either the Bayesian learning community or the NLP
community, please complete our survey (available on the web page),
which will serve to guide the panel discussions at the workshop.


ORGANIZATION
------------

Hal Daume III
Information Sciences Institute
hdaume at isi.edu
http://www.isi.edu/~hdaume/

Yee Whye Teh
National University of Singapore
tehyw at comp.nus.edu.sg
http://www.cs.berkeley.edu/~ywteh/