2012 IEEE WCCI

Special Session on Nonnegative Matrix factorization paradigm for unsupervised learning

 

NEWS

Paper Submission Deadline has been extended to Jan 18, 2012 (23:59 EST) - The Paper Submission System will close straight after that, no more extensions will be given.  

AIMS  AND SCOPE

This special session will cover original and pioneering contributions, theory as well as applications on nonnegative matrix factorization (NMF) paradigm for unsupervised learning, and aim at an inspiring discussion on the recent progress and the future development.
A fundamental problem in many machine learning tasks is to find a suitable representation of the data. A useful representation typically makes latent structure in the data explicit, and often reduces the dimensionality of the data so that further computational methods can be applied.
NMF is a commonly used approach to understanding the latent structure of the observed matrix for various applications. NMF methods have attracted increasing attention in recent years because of their mathematical elegance and encouraging empirical results.
There are many forms of NMF. Previous work has shown that by respecting the nonnegativity, the factorization results will be easier to interpret while being comparable to, or better than, other techniques like SVD on effectiveness NMF has been successfully applied to a variety of applications, including face detection and recognition, audio and speech processing, text mining, biomedical image analysis, bioinformatics, and so on.

In this special session, the main methods of matrix factorization paradigm for unsupervised learning will be presented. Also, the effectiveness of these methods will be discussed considering the concepts of diversity and selection of these approaches.

Topics of interest include but not limited to:
    - Convex-NMF
       - Hard clustering and NMF
       - Kernel-NMF
       - NMF for Large-Scale Data
       - Maximum margin matrix factorization (MMMF)
       - NMF with Sparseness Constraints
       - Orthogonal symmetric NMF
       - Probabilistic NMF
       - Relaxed NMF
       - Semi-NMF
       - Tri-NMF
       - Weighted NMF
       - Weighted NMTri-Factorization
       - Dimensionality reduction via matrix factorization