IEEE Signal Processing Magazine Special Issue on: Dimensionality Reduction via Subspace and Manifold Learning

IEEE Signal Processing Magazine Special Issue on:
Dimensionality Reduction via Subspace and Manifold Learning

Call For Papers



Scope and Purpose:

The problem of finding and exploiting low-dimensional structures in high-dimensional data is taking on increasing importance in image, video, or audio processing, web data analysis/search, and bioinformatics, where datasets now routinely lie in thousands to millions-dimensional observation spaces. The curse of dimensionality is in full play here: We often need to conduct meaningful inference with limited number of samples in a very high-dimensional space. Conventional statistical and computational tools are often severely inadequate for processing and analyzing high-dimensional data. Although the data might be presented in a high-dimensional space, their intrinsic complexity and local dimensions are typically much lower.

This special issue of IEEE Signal Processing Magazine is to attract articles that cover existing approaches to dimension reduction based on learning of subspaces or submanifolds -- from linear to nonlinear models, from homogeneous to hybrid models, from statistical, to geometric, to algebraic, and to graphical methods. We would also like to feature many successful applications of these new methods, including but not limited to signal/image processing, pattern recognition, bioinformatics, and web data mining. Below is an incomplete list of potential topics to be covered in the special issue:


Important Dates (timetable updated!):


Manuscript Submission Guidelines:

Invitations to full article have been sent out to authors. Please make sure you submit your full aritcle for review in about 60 days after receiving the notice. We have invited for 15 full articles (out of 68 whitepapers), but that is more than the final issue can accomondate. So the invitation does not mean acceptance! The submitted articles will still undergo independent, blind peer review for final selection. If you have any concern about this, please contact the editors. The selection criteria are mostly based on the following two aspects:

Paper Format:

You must prepare the manuscript in single-space double-column format. Your manuscript must be 8 pages or shorter. There is no template available since the IEEE Magazine uses a different format from IEEE Transactions. Our suggestion is that you use the single-space double-column format from the following conference template (http://icassp09.com/papers/PaperKit.html#Templates). It is not the final format, but it is close enough to give you an idea about the page limits. In the final SP Magazine format, the manuscript will be about 1 page longer or 10% longer after typesetting if the authors submitted the original manuscript in the ICASSP format.

Prospective authors should submit their manuscripts to the web submission system through IEEE Manuscript Central at: http://mc.manuscriptcentral.com/spmag-ieee.



Guest Editors:

Lead Guest Editor:
Associate Guest Editors: All manuscripts will undergo independent, blind peer reviews, administered by all the editors. Acceptance will be based on the quality of the paper and its relevance to the special issue. If you have specific questions about the review and decision of your manuscript, please contact the Chief Editor, Prof. Yi Ma: yima@uiuc.edu.


©2008 Yi Ma
Last modified: Sat Oct 18 16:53:46 CDT 2008