[Comp-neuro] PRL Special Issue on Partially Supervised Learning

friedhelm.schwenker at uni-ulm.de friedhelm.schwenker at uni-ulm.de
Wed Feb 15 23:03:00 CET 2012

Call for Papers

Special Issue on

Partially Supervised Learning for Pattern Recognition

to be published in  Pattern Recognition Letters

*** Submission deadline: June 30, 2012 ***

Partially supervised learning (PSL) is a general framework for  
learning with labeled and unlabeled data. In traditional pattern  
classification a label (the correct class) is associated with each  
training pattern; in the PSL framework this label might as well be  
crisp, but it might also be paired with a confidence value, or it  
might be an imprecise and/or uncertain soft label (defined through  
certain types of uncertainty models), or it might be that such a label  
is not available at all.

PSL thus generalizes, involves, or builds upon several kinds of  
learning paradigms that have also found application to pattern  
classification problems. Such paradigms include: supervised and  
unsupervised techniques; semi-supervised learning; transductive,  
transfer, and diffusion learning; policy learning in partially  
observable environments. Therefore PSL methods and algorithms for  
pattern recognition are of great interest in both practical  
applications and theory. Research in the field of PSL is still in its  
early stages and has great potential for further growth.

This special issue invites paper submissions on the most recent  
developments in PSL research rooted in (or, aimed at) pattern  
recognition. The special issue will comprise (1) papers submitted in  
response to this call, and (2) extended versions of selected papers  
from the recent, successful PSL 2011 Workshop held in Ulm (Germany)  
(http://neuro.informatik.uni-ulm.de/PSL2011/), sponsored by the  
International Association for Pattern Recognition.

Topics of interest include (yet, they are not limited to) the  
following issues.

Methodological issues (as long as they relate to pattern recognition:
•	Combinations of supervised and unsupervised learning
•	Diffusion learning
•	Semi-supervised classification and clustering
•	PSL with deep architectures
•	Active leaning
•	PSL with vague, fuzzy, or uncertain teaching signals
•	PSL in multiple classifier systems and ensembles
•	PSL in neural nets, machine learning, or statistical pattern recognition
•	Transfer learning
•	Transductive learning

Pattern recognition applications of PSL in:
•	Image and signal processing
•	Multimodal information processing
•	Information fusion
•	Data mining and web mining
•	Bioinformatics/Cheminformatics

Paper submission
Papers must be submitted online via the Pattern Recognition Letters  
website (http://ees.elsevier.com/patrec/), selecting the choice that  
indicates this special issue (identifier: PSL-PR). Prepare your paper  
following the Journal guidelines for Authors  
(http://www.elsevier.com/wps/find/journaldescription.cws_home/505619/authorinstructions), which include specifications for submissions aimed at Special Issues. In particular, a maximum of 7500 words is admitted for special issue papers, without counting the References (plus at most 10 Figures/Tables in total).  Priority will be given to the papers with high novelty and  

*** Submission deadline: June 30, 2012 ***

(note: electronic submission opens on May 30, 2012)

If you are not sure on whether your manuscripts matches the aims and  
scope of this special issue or not, do not hesitate to get in touch  
with the guest editors at any time.

Guest editors
Friedhelm Schwenker, University of Ulm, Germany  
(friedhelm.schwenker at uni-ulm.de)
Edmondo Trentin, University of Siena, Italy  (trentin at dii.unisi.it)

More information about the Comp-neuro mailing list