WPOS 2012 - NIPS 2012 Workshop on Perturbations, Optimization, and Statistics

NIPS 2012 Workshop on Perturbations, Optimization, and Statistics
December 7 or 8 (TBD), 2012 at Lake Tahoe, Nevada, U.S.A.
Web Site: http://www.stat.ucla.edu/~gpapan/pos12/
Submission Deadline: September 30, 2012
============================
== Overview ==
In nearly all machine learning tasks there exists randomness, or noise, in the
observed data and in the relationships encoded by the model. Often this noise
is considered undesirable and would be eliminated if possible. However, there
is an emerging body of work on perturbation methods, showing the benefits of
modeling, learning, and inference pipelines that explicitly consider the
addition of noise. Perturbation methods provide a principled way to reason
about neighborhoods of possible outcomes when trying to make a decision which,
for example, is robust to small changes in model parameters or needs to
account for a lack of knowledge by averaging over different possible
outcomes. Recently, several works influenced by diverse fields of research
such as statistics, optimization, machine learning, and theoretical computer
science, have used perturbation methods in similar ways. The workshop aims to
bring together the growing community of researchers interested in various
aspects of this area, to broaden our understanding of why and how perturbation
methods can be useful, and to explore applications of these methods in areas
such as computer vision and language modeling.
== Call for Papers ==
In addition to a program of invited presentations, we solicit contribution of
short papers that explore perturbation-based methods in the context of topics
such as: statistical modeling, sampling, inference, estimation, theory, robust
optimization, robust learning. We are interested in both theoretical and
application-oriented works. We also welcome papers that explore connections
between alternative ways of using perturbations.
Contributed papers should adhere to the NIPS format and be up to four pages
long (without counting the list of references). Papers submitted for review do
not need to be anonymized. There will be no official proceedings. Thus, apart
from papers reporting novel unpublished work, we also welcome submissions
describing work in progress or summarizing a longer paper under review for a
journal or conference (this should be clearly stated though). Accepted papers
will be presented as posters; some may also be selected for spotlight talks.
Please submit papers in PDF format by email to posNIPS2012-AT-gmail.com. The
submission deadline is September 30, 2012 and notifications of acceptance will
be sent by October 14, 2012. At least one of the authors must be attending the
workshop to present the work.
== Confirmed Invited Speakers ==
Ryan Adams (Harvard), Amir Globerson (HUJI), Tommi Jaakkola (MIT), Andrea
Montanari (Stanford), Yuval Peres (Microsoft Research), Pascal Vincent
(Montreal), Max Welling (UCI), Alan Yuille (UCLA).
== Organizers ==
Tamir Hazan (TTI-C), George Papandreou (UCLA), Danny Tarlow (Univ. of Toronto).