Activity Update from the MLSP TC

Contributed by Kenneth E. Hild II (TC Liaison to eNews)

The Machine Learning for Signal Processing Technical Committee (MLSP TC) continues to be very active. Below I summarize some of the items we have been working on during the past five months including our: upcoming MLSP workshop, upcoming Data Analysis Competition, recently elected/nominated/retiring MLSP members, upcoming MLSP TC meeting, and recent MLSP awards.

MLSP Workshop

Everyone is invited to attend the next MLSP workshop (Conference id # 19730), which will be held at The Royal Palace of the Magdalena in Santander, Spain from September 23-26. The Palace is located on a peninsula, which is surrounded by trees and beaches and offers spectacular views of the Bay of Santander. This will mark the 22nd annual meeting organized by the MLSP TC. At the workshop, you can expect to find and learn about the most recent and exciting advances in data analysis for signal processing by means of tutorials, keynote talks, and both special and regular single-track sessions. Keynote lectures will be given by: Martin Wainwright, Francis Bach, and Ali H. Sayed. Topics for the workshop include, but are not limited to: Learning Theory, Graphical Models, Pattern Recognition and Classification, Distributed, Bayesian, Subspace, and Sparsity-Aware Learning, Multi-modal Data Fusion, Perceptual Signal Processing, and Cognitive Information Processing. Applications include: speech, audio, video, music, biomedical imaging, communication, bioinformatics, computational genomics, social networks, biometrics, energy, and smart grid. Don't miss the deadline for submission of full papers, which is May 7, 2012.

Also, mark on your calendars for the 2013 MLSP Workshop, which will be located in Istanbul, Turkey. The location of the 2014 MLSP Workshop will be determined soon. The two finalists for 2014 are Reims, France and Daejeon, Korea.

Data Analysis Competition

For several years now, the MLSP TC has hosted an annual Data Analysis Competition. The competition is open to everyone. This year the goal is to select/design a classifier and any necessary pre-processing system that correctly classifies an employee/resource pair (a resource can be, e.g., a computer, data base, or data portal) into one of two classes: access or non-access. The winner will be the submission that minimizes the number of manual access grant/removal operations in the disjoint and unpublished testing data. Up to three Kindle Fire tablet computers from Amazon will be awarded as prizes. Amazon will also invite the winners to present their results at Amazon's new SLU headquarters in Seattle, Washington, USA. For more details, refer to the Vol. 28, No. 6, November, 2011, pp. 193-196).

In the coming months, the MLSP TC will be reviewed by the Signal Processing Society at ICASSP as part of their standard review process. We would like to thank Tulay Adali for her efforts in compiling the fairly-length report that is associated with the review.

In February of this year, the new TSP EDICS submitted by the MLSP TC was approved by the IEEE Publications Board. The most important additions are the inclusion of the following application areas: Applications in Distributed and Cooperative Learning, Applications in Music and Audio Processing, Applications in Data Fusion, and Applications in Systems Biology.

Awards

Finally, we wish to congratulate Tulay Adali, who was recently named a 2012 IEEE Signal Processing Society Distinguished Lecturer. Tulay, who was nominated by the MLSP TC, is one of only five Distinguished Lecturers appointed this year. She will travel around the world and present her research on data-driven and complex-valued signal processing and their applications to medical image analysis.