September 2011 Archives

As a technical program chair, I am happy to announce the first call for papers and works for the 9th Sound and Music Computing Conference, 12-14 July 2012 Medialogy section,  Department of Architecture, Design and Media Technology, Aalborg University Copenhagen
http://smc2012.smcnetwork.org/. (And, of course, I especially encourage the submission of work that uses principles of sparsity :)
 
The SMC Conference is the forum for international exchanges around the core interdisciplinary topics of Sound and Music Computing, and features workshops, lectures, posters, demos, concerts, sound installations, and satellite events. The SMC Summer School, which takes place just before the conference, aims at giving young researchers the opportunity to interactively learn about core topics in this interdisciplinary field from experts, and to build a network of international contacts. The specific theme of SMC 2012 is "Illusions", and that of the SMC Summer School is "Multimodality".

================Important dates=================
Deadline for submissions of music and sound installations: Friday, February 3, 2012
Deadline for paper submissions: Monday 2 April, 2012
Notification of music acceptances: Friday, March 16, 2012
Deadline for applications to the Summer School: Friday March 30, 2012
Notification of acceptance to Summer School: Monday April 16, 2012
Deadline for submission of final music and sound installation materials: Friday, April 27, 2012
Notification of paper acceptances: Wednesday 2 May, 2012
Deadline for submission of camera-ready papers: Monday 4 June, 2012
SMC Summer School: Sunday 8 - Wednesday morning 11 July, 2012
SMC Workshops: Wednesday afternoon 11 July, 2012
SMC 2011: Thursday 12 - Saturday 14 July, 2012
===========================================

SMC2012 will cover topics that lie at the core of the Sound and Music Computing research and creative exploration.
 We broadly group these into:
  - processing sound and music data
  - modeling and understanding sound and music data
  - interfaces for sound and music creation
  - music creation and performance with established and novel hardware and software technologies

================Call for papers==================
SMC 2012 will include paper presentations as both lectures and poster/demos. We invite submissions examining all the core areas of the Sound and Music Computing field. Submission related to the theme "Illusions" are especially encouraged. All submissions will be peer-reviewed according to their novelty, technical content, presentation, and contribution to the overall balance of topics represented at the conference. Paper submissions should have a maximum of 8 pages including figures and references, and a length of 6 pages is strongly encouraged. Accepted papers will be designated to be presented either as posters/demos or as lectures. More details are available at http://smc2012.smcnetwork.org/
===========================================

=======Call for music works and sound installations========
SMC 2012 will include four curated concerts addressing the conference topic "Illusions". We invite submissions of original compositions created for acoustic instruments and electronics, novel instruments and interfaces, music robots, and speakers as sound objects. Submissions of sound installation are also encouraged. See curatorial statements and call specifics at: http://smc2012.smcnetwork.org.
==============================================================

Phase transitions for MP?

| No Comments
At SPARS 2011, I had a conversation that confused me. MP has no phase transition, or, more accurately, the phase transition is at a sparsity of zero. In other words, as the ambient dimension \(N\) grows, the sparsity at which MP successfully recovers a compressively sensed sparse vector goes to zero. However, that is not what I am finding empirically. In the image below you can see the phase transitions for "MP+" (with a debiasing step), for several sparse vector distributions, for \(N=800\). (B = Bernoulli, U = Uniform, L = Laplacian, N = Normal, BR = Bimodal Rayleigh.) The criteria for successful recovery is that the solution support is exactly the same as the original sparse vector. For each pair of sparsity and indeterminacy I run 100 trials. (Notice the vertical scale.)

MP+.png Just to be sure my sinc interpolation scheme is not causing problems, I also implemented the approach by Maleki and Donoho, which estimates the phase transition by fitting a linear model in \(\rho\) to the logit of the estimated success probability. These phase transitions are the dotted lines. For the most part, the pairs line up, though it appears at times the interpolation approach is too optimistic for bimodal Rayleigh.

Of course, why would one use the pure greedy algorithm when its cousin PrOMP does so much better? Or IHT? I remember during his talk at SPARS2011, Jeffrey Blanchard said that if one knows where in the phase plane a problem lies, then we can find the least-cost algorithm to solve it. And MP is a pretty low-cost algorithm.

Anyhow, I find it extremely strange that MP, the king of pure greedy algorithms, does so poorly for sparse vectors distributed in ways that are usually favored by greedy approaches. And its performance gets worse in these cases with an increasing number of measurements. In such cases, the coherence of the dictionary should be decreasing, and so I would think MP would be at more of an advantage.

I am now running the same experiments for \(N=2000\) to see how things change.

Update 20111004: Not much changes at the higher dimension.
Hello, and welcome to Paper of the Day (Po'D): Precise Undersampling Theorems Edition. Today's paper happens to have the same title! D. L. Donoho and J. Tanner, "Precise undersampling theorems," Proc. IEEE, vol. 98, no. 6, pp. 913-924, June 2010.

MMSE estimators of mean and variance

| 1 Comment
I am at that point in preparing my lecture notes where I am confusing myself, i.e., done with the elementary things, and into the material I haven't studied for many years. Let's say we have a set of \(N\) observations \(\{z_n\}\) of some random variable \(Z\). We don't know how \(Z\) is truly distributed, but we can estimate the mean and the variance of the observations, so that we might, e.g., model it by a Gaussian distribution function.

Blog Roll

About this Archive

This page is an archive of entries from September 2011 listed from newest to oldest.

August 2011 is the previous archive.

October 2011 is the next archive.

Find recent content on the main index or look in the archives to find all content.