Thursday, April 26, 2012

GraphLab 2012 Workshop registration and other random items.

Danny Bickson just sent me the following about regsitration for the upcoming GraphLab workshop (see GraphLab workshop, Why should you care ?A small Q&A with Danny Bickson on GraphLab ).:

As you may be aware, based on your feedback, we setup the workshop data to Monday July 9th in San Francisco. The workshop will take place one day before Stanford MMDS workshop, thus allowing people to attend both workshops.

We have finally setup the workshop website: Early registration is now open.

At this point, we would like your help in disseminating the news about the event to anyone relevant.

Thanks a lot for your great help! Looking forward to meeting everyone in person at the workshop!

Dr. Danny Bickson
Project Scientist, Machine Learning Dept.
Carnegie Mellon University
As I mentioned to one of you recently, I have scheduling conflicts and won't probably attend but I think it is a good opportunity to invest in a plane ticket and attend. It would even be better if you could have a presentation or a poster. Let me or Danny know if you have a poster or a presentation idea in mind. The presentation and posters currently on the program are listed here. The companies represented there are listed here and here. (of interest: GAMP: Generalized Approximate Message Passing in GraphLab )

Yesterday's mention of the paper that appeared in  Nature: Faster STORM using compressed sensing by Lei Zhu, Wei Zhang, Daniel Elnatan & Bo Huang and for which there is no preprint available, got me to find these two items:

* One on PSF engineering and STORM. Eventually some of that has to spill to compressive sensing methods Optimal 3D single-molecule localization for superresolution microscopy with aberrations and engineered point spread functions by Sean QuirinSri Rama Prasanna Pavani, and Rafael Piestun. The abstarct reads:

Photo-activation localization microscopy is a far-field superresolution imaging technique based on the localization of single molecules with subdiffraction limit precision. Known under acronyms such as PALM (photo-activated localization microscopy) or STORM (stochastic optical reconstruction microscopy), these techniques achieve superresolution by allowing only a sparse, random set of molecules to emit light at any given time and subsequently localizing each molecule with great precision. Recently, such techniques have been extended to three dimensions, opening up unprecedented possibilities to explore the structure and function of cells. Interestingly, proper engineering of the three-dimensional (3D) point spread function (PSF) through additional optics has been demonstrated to theoretically improve 3D position estimation and ultimately resolution. In this paper, an optimal 3D single-molecule localization estimator is presented in a general framework for noisy, aberrated and/or engineered PSF imaging. To find the position of each molecule, a phase-retrieval enabled maximum-likelihood estimator is implemented. This estimator is shown to be efficient, meaning it reaches the fundamental Cramer–Rao lower bound of x, yand z localization precision. Experimental application of the phase-retrieval enabled maximum-likelihood estimator using a particular engineered PSF microscope demonstrates unmatched low-photon-count 3D wide-field single-molecule localization performance.

* And the other one by Roarke Horstmeyer on Measuring Ultrafast Pulses: Frequency Resolved Optical Gating and Phase Space Functions, 6.638 Ultrafast Optics Final Project, Dec. 2010 . At the very end, NMF is used to perform a FROG deconvolution for certain types of pulses. We've talked about FROG before (see also Shouldn't We Call It "Sparse Sensing" or "Sparsity Sensing" instead ? )

Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.

No comments: