Thursday, 19 June 2008

More MIR related PhDs

I just updated the list of MIR PhDs. The most notable update is the thesis by Adam Berenzweig. Unfortunately his thesis is not publicly available for download, but it seems like he has found an answer to why we've been observing hubs using certain similarity measures for music. Quoting from his abstract: "A practical problem with this technique, known as the hub phenomenon, is explored, and we conclude that it is related to the curse of dimensionality."

I'm sure there's still plenty of dissertations missing in the list and I'll be happy to add any that are sent to me. (I'll also be happy to update any broken links or missing information...)

Btw, the list is slowly approaching 100 entries now. The Matlab script I wrote to generate the html files and statistics now takes almost 5 seconds to complete.


Foafing your music said...

(NB I'm sure you're aware of these papers, but I couldn't resist, as I really like 'em :-)

Regarding hubs in audio similarity, I really like JJ's paper:
Aucouturier, J.-J. and Pachet, F. A scale-free distribution of false positives for a large class of audio similarity measures. Pattern Recognition, vol. 41(1), pp. 272-284, 2007.

Also Tim's work is very interesting. E.g. Automatically Adapting the Structure of Audio Similarity Spaces

And, yes, I'm a bit skeptical of pure CB measures so, for me, these kind of papers hit the nail in the head.

Cheers, Oscar

jeremy said...

I thought Adam went to Google back in 2003 or 2004? Did he leave Google and go back to his PhD? The date on the PhD is 2007.

And Oscar: I share your skepticism of pure CB measures. At the same time, I am also skeptical of measures than have zero CB component at all! ;-)

Elias said...

@Oscar: I totally agree with that selection of references... and that's a perfect "hub" picture of Tim :-)

@Jeremy: I have no clue what Adam is up to. I assumed he had continued working for Google while finishing his thesis...