Thursday, November 10, 2011

The problem with (pre-publication) peer-review

Using materials from the previous two posts.

Original figure and computations: Bob SturmCyclic Greedy Algorithms for Recovering Compressively Sampled Sparse Signals

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.


Bob et Carla said...

I don't know whether the Mallat and Zhang MP article made it through peer review on the first try. It was essentially one of many reinventions of an idea proposed long before, but now presented to the signal processing community (again).

Do you know why the SL0 article was rejected at first? It is published in 2009 IEEE Trans. Sig. Proc. And before that it was presented at the 2007 ICA.

Igor said...

I probably need to ask the authors but the solver became known with 2007/2008 papers which led me to ask the authors to put it on the web.

My generic complaint is simple, however one wants to see the peer review process as is currently being implemented it fails in one basic tenet: it is not a review as one cannot judge if an algo is any good based if the "reviewer" does not have access ot it.