Activity Update from the MLSP TC

by Kenneth E. Hild II (MLSP Liaison to E-Newsletter)

The most recent Machine Learning for Signal Processing (MLSP) workshop was held on September, 2011 in the National Laboratory of Pattern Recognition (NLPR), Institute of Automation, Chinese Academy of Sciences, which is located in Beijing, China. The workshop was a huge success. We received 134 submissions, 96 of which (72%) were accepted and included in the proceedings. Of the 96 accepted papers, 35 were orally presented and 61 were presented as posters. We also hosted a Face Recognition competition. The entrant with the highest performing algorithm authored an invited article, which we subsequently published in the proceedings, describing the winning method. Popular topics in this year's program included dimensionality reduction, information visualization, neuro-informatics, non-negative matrix factorization, kernels, Gaussian processes, and dynamical state-space models. We also published papers discussing signal, image, video, and sensory data processing.

The next MLSP workshop will be held on September 23-26, 2012 and will be located at the beautiful Royal Palace of the Magdalena in Santander, Spain. Full paper submission deadline is May 7, 2012.

Selected papers from MLSP 2012 will be considered for a special issue of  the International Journal of Neural Systems (JCR IF 4.2). We plan to once again host a data analysis competition and to award prizes to as many as the top three applicants (the awards in the past have included Nokia smart phones and travel grants). Prospective authors are invited to submit 6-page papers on relevant algorithms and applications including, but not limited to:

• Learning theory and techniques;
• Graphical models and kernel methods;
• Data-driven adaptive systems and models;
• Pattern recognition and classification;
• Distributed, Bayesian, subspace/manifold, and sparsity-aware learning;
• Multiset data analysis and multimodal data fusion;
• Perceptual signal processing in audio, image and video;
• Cognitive information processing;
• Applications, including: speech, audio and video, music processing, biomedical imaging, communications, bioinformatics, computational genomics, social networks, biometrics, energy and smart grid.

Invited speakers include Martin Wainwright (Univ. California at Berkeley, USA), Francis Bach (INRIA, France), and Ali H. Sayed (Univ. California at Los Angeles, USA).

More information about the upcoming MLSP conference in 2012, as it becomes available, can be obtained on the conference website.

On a different note, the IEEE Signal Processing Society recently announced that the current chair of the MLSP TC, Tülay Adali, will be one of only five Distinguished Lecturers for 2012 and 2013. She will be giving talks on the following five topics:

1. Data-driven Analysis and Fusion of Medical Imaging Data;
2. Complex-valued Adaptive Signal Processing: When and How to Take Noncircularity into Account;
3. ICA, ISA, and IVA: Theory, Connections, and Applications in Medical Image Analysis;
4. Optimization in the Complex Domain using Wirtinger Calculus: Applications to ICA;
5. Joint Blind Source Separation: Applications in Medical Image Analysis.

Also, another one of the MLSP TC members, Gustavo Camps-Valls, has recently published a book [1], was awarded the Best Paper award of the IEEE Geoscience and Remote Sensing Society 2011 [2], and another one of his papers [3] was identified by ISI/ThomsonReuters as a Fast Moving Front research. Science Watch conducted the interview associated with this latter award. We wish to congratulate both Tulay and Gustavo on their recent accomplishments.

[1] "Remote Sensing Image Processing", G. Camps-Valls, D. Tuia, L. Gomez-Chova, and J. Malo, editors. Collection ‘Synthesis Lectures on Image, Video, and Multimedia Processing’, Al Bovik, Ed., Morgan & Claypool Publishers, LaPorte, CO, USA, Sept 2011.
[2] "Remote Sensing Feature Selection by Kernel Dependence Estimation", G. Camps-Valls, J. Mooij and B. Schölkopf IEEE Geoscience and Remote Sensing Letters, 7(3), 587-591, July 2010
[3] "Kernel-based Methods for Hyperspectral Image Classification", G. Camps-Valls and L. Bruzzone IEEE Transactions on Geoscience and Remote Sensing, 43(6), 1351-1362.