<html>
<head>
<meta http-equiv="content-type" content="text/html;
charset=ISO-8859-1">
</head>
<body text="#000000" bgcolor="#FFFFFF">
[Apologies if you receive multiple copies of this CFP]<br>
<br>
<b>Call for papers: special session on "</b><b><b>Label noise in
classification</b>" at </b><b>ESANN 2014</b><br>
<br>
European Symposium on Artificial Neural Networks, Computational
Intelligence and<br>
Machine Learning (ESANN 2014). 23-25 April 2014, Bruges, Belgium - <a
class="moz-txt-link-freetext" href="http://www.esann.org">http://www.esann.org</a><br>
<span style="font-family: Verdana; font-size: 13.5pt; line-height:
20px; "><br>
<br>
</span><b>DESCRIPTION</b>:<br>
<br>
In classification, it is difficult to obtain completely reliable
labels. Indeed, labels are often polluted by label noise, due to
e.g. insufficient information or expert mistakes. Many works have
tackled the problem of learning in the presence of label noise.
Filtering methods have been developed to detect and remove
mislabelled instances. Also, recent approaches attempt to take
label noise into account while learning, using e.g. probabilistic
models of label noise or prior knowledge about the influence of
label noise on specific methods. Other settings like e.g.
semi-supervised learning have also been studied.<br>
<br>
This special session aims to provide a forum where researchers could
discuss the most recent developments in the field of label noise.
Contributions should propose new methods to deal with label noise,
new applications where label noise must be taken into account,
theoretical results about learning in the presence of label noise or
experimental results which provide insight about existing methods.<br>
<br>
Examples of topics of interest include (but are not limited to) the
following:<br>
<ul>
<li>when are noisy labels better than no labels at all?</li>
<li>what makes a classifier robust to label noise?</li>
<li>dealing with different types of label noise (random,
non-random, malicious, or adversarial)</li>
<li>conditions for the consistency of classification in the
presence of label noise</li>
<li>label noise in high dimensional small sample settings</li>
<li>the issue of model order selection in the presence of label
noise</li>
<li>feature selection and dimensionality reduction in the presence
of label noise</li>
<li>label-noise aware classification algorithms in static and
dynamic scenarios</li>
<li>learning with side information to counter label noise</li>
</ul>
<b><br>
<br>
SUBMISSION:<br>
</b><br>
Prospective authors must submit their paper through the ESANN portal
following the instructions provided in <a
class="moz-txt-link-freetext"
href="http://www.elen.ucl.ac.be/esann/index.php?pg=submission">http://www.elen.ucl.ac.be/esann/index.php?pg=submission</a>.
Each paper will undergo a peer reviewing process for its
acceptance. Authors should send as soon as possible an e-mail with
the tentative title of their contribution to the special session
organisers.<b><br>
<br>
<br>
IMPORTANT DATES</b><b>:</b><br>
<br>
Paper submission deadline : 29 November 2013<br>
Notification of acceptance : 31 January 2014<br>
The ESANN 2014 conference : 23-25 April 2014<br>
<b><br>
<br>
SPECIAL SESSION ORGANISERS</b><b>:</b><br>
<br>
Dr. Benoît Frénay<br>
Université catholique de Louvain, Belgium<br>
E-mail: <a class="moz-txt-link-abbreviated"
href="mailto:benoit.frenay@uclouvain.be">benoit.frenay@uclouvain.be</a><br>
Phone: +32 10 478133<br>
<br>
Dr. Ata Kaban<br>
University of Birmingham, United Kingdom<br>
E-mail: <a class="moz-txt-link-abbreviated"
href="mailto:A.Kaban@cs.bham.ac.uk">A.Kaban@cs.bham.ac.uk</a><br>
Phone: +44 121 41 42792
</body>
</html>