Call for Paper


The theme of the workshop


Partially supervised learning (PSL) is a rapidly evolving area of machine learning. In many applications unlabeled data may be relatively easy to collect, whereas labeling this data is difficult, expensive or/and time consuming as it needs the effort of human experts. PSL is a general framework for learning with labeled and unlabeled data, for instance in classification, it is assumed that each learning sample consists of a feature vector and some information about its class. In the PSL framework this information might be a crisp label, or a label plus a confidence value, or it might be an imprecise and/or uncertain soft label defined through certain type of uncertainty model (fuzzy, Dempster-Shafer), or it might be that information about a class label is not available.

The PSL framework thus generalizes many kinds of learning paradigms including supervised and unsupervised learning, semi-supervised learning for classification and regression, transductive learning, semi-supervised clustering, policy learning in partially observable environments, and many others. Therefore PSL methods and algorithms 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, thus, leaving plenty of room for further development.

This 1st workshop on Partially Supervised Learning endeavors to bring together recent novel research in this area and to provide a forum for further discussion.


Topics of interest include, but are not limited to:

Methodological issues:

  • Combinations of supervised and unsupervised learning.
  • Deep learning
  • Semisupervised learning in feed forward networks and kernel machines, recurrent and competitive neural networks.
  • Learning with vague, fuzzy, or uncertain teaching signals.
  • Semisupervised clustering.
  • Learning with reward.
  • Semisupervised classification in multiple classifier systems and ensembles.

Applications:

  • Image processing and segmentation
  • Sensor and information fusion
  • Multimodal information processing.
  • Feature extraction, dimension reduction, clustering and vector quantization
  • Speech and speaker recognition.
  • Human computer interaction
  • Data mining
  • Bioinformatics.


Original and unpublished contributions are solicited which include regular papers and extended abstracts. Maximum paper length for regular papers is 10 pages (4 pages for extended abstracts) in LNCS/LNAI format. Proceedings will be published as a volume of the Springer LNAI series. Instructions for authors, LaTeX templates,etc are available at the Springer LNCS/LNAI website. Submission of a paper constitutes a commitment that, if accepted, one or more authors will attend the workshop.

Important dates

Paper submission: May 6, 2011 extended to July 15, 2011
Notification of acceptance: August 1, 2011
Camera ready version: October 15, 2011

Workshop : September 15-16, 2011
Publication of proceedings : December 2011

Comments are closed.