Comments for regularize

Comment on Comparing shapes: From Prokhorov/Wasserstein over Hausdorff to Gromov-Hausdorff metrics by Comparing shapes: metric measure spaces | regularize

Comparing shapes: metric measure spaces | regularize — Tue, 19 Mar 2013 10:52:08 +0000

[...] I continue my previous post on methods to compare shapes. In that post we started with different metrics between two probability measures and defined on [...]

Comment on A mathematical social network? by Dirk

Dirk — Sun, 17 Feb 2013 19:02:05 +0000

If the AMS really provides a social network, this would be a good reason for me to join. Nice interfaces to other databases and services (e.g. arXiv, MathSciNet, maybe even mathoverflow of other stackexchange sites,…) seem to be obvious. Also possibilities for joint calenders and events should be there. Tools for online collaboration would also be great (e.g. in spirit of Mendeley). I think one could come up with a fairly long list of wishes easily…

Comment on A mathematical social network? by Peter Krautzberger

Peter Krautzberger — Sun, 17 Feb 2013 00:10:08 +0000

Thanks for the trackback. I agree that an international effort would be awesome, but I do understand the need to isolate and provide benefits only to a society’s members. That’s why I’d go with a federated tool like Diaspora where different “pods” (that’s what servers are called) could connect. It shouldn’t be hard to have extra features just for society members.

From futher rumors, it seems the AMS is taking this seriously (though unfortunately not in the federated direction). Question is: what are the extra benefits? An easy one would be MathJax support. But I think the key is to redefine what our societies are — from discrete to continuous connectivity, if you will.

It irks me a little that the members are not involved in determining this…

Comment on Socializing the German math community? by A mathematical social network? « regularize

A mathematical social network? « regularize — Thu, 14 Feb 2013 07:49:15 +0000

[...] of the AMS, the DMV, already has a default social network for all its members (I blogged on this in an earlier post). This is pretty hard to find on the DMV homepage and it seems that almost nobody is using it for [...]

Comment on Part time mathematician for a year by Kanglin

Kanglin — Mon, 19 Nov 2012 21:48:24 +0000

Worth knowing! Thank you, Dirk!

Comment on What happens when you get a “Ruf” (i.e. a job offer) by Part time mathematician for a year « regularize

Part time mathematician for a year « regularize — Mon, 19 Nov 2012 20:04:58 +0000

[...] dates (e.g. the ones related with job interviews and negotiations on which I have written here and here). In these cases, I also mentioned my limited possibilities for meetings and the severe [...]

Comment on What they’ll ask in the job interview by Part time mathematician for a year « regularize

Part time mathematician for a year « regularize — Mon, 19 Nov 2012 20:04:56 +0000

Comment on What is convexity? by Mehmet Suzen

Mehmet Suzen — Thu, 15 Nov 2012 21:49:37 +0000

Nice tutorial. Convexity is vital in Legendre-Fenchel transforms in statistical mechanics. Probably it is worth to mention Jensen’s Inequility as well. Quite good introduction is from Needham (“A Visual Explanation of Jensen’s Inequality”, American Mathematical Monthly 100(8):768–71.)

Comment on What they’ll ask in the job interview by Dirk

Dirk — Wed, 31 Oct 2012 11:27:23 +0000

I think “big shot” translates to “Hohes Tier” in german. What I meant is that the committee will ask highly reputbale mathematicians in the respective field. Sorry for the jargon…

Comment on What they’ll ask in the job interview by Alexander

Alexander — Wed, 31 Oct 2012 10:52:43 +0000

Dirk, what does it mean “and ask some big shots in the field for reports” ?

Comment on Leftovers in my RSS reader by Dirk

Dirk — Sat, 06 Oct 2012 13:40:48 +0000

It’s a Asus Zenbook Prime. Looks cool, feels cool and works ok with Ubuntu (Touchpad control could be better…).

Comment on Leftovers in my RSS reader by Andreas Bartels

Andreas Bartels — Sat, 06 Oct 2012 09:58:52 +0000

Hi Dirk! What’s your new laptop?

Comment on What they’ll ask in the job interview by What happens when you get a “Ruf” (i.e. a job offer) « regularize

Tue, 02 Oct 2012 20:29:41 +0000

[...] a previous post I wrote about my experience with three job interviews for math professorships in Germany I had [...]

Comment on Semi-continuous sparse reconstruction and compressed sensing by The “simplex of probabilty measures” and semi-continuous compressed sensing « regularize

Thu, 20 Sep 2012 09:18:25 +0000

[...] that something similar to “support-pursuit” does not work here: The minimization problem does not make much sense, since for all [...]

Comment on No, we couldn’t. Why the DMV did not change its name? by Dirk

Dirk — Mon, 17 Sep 2012 19:38:18 +0000

Sounds good! But probably several people in the DMV will have some objection…

Comment on No, we couldn’t. Why the DMV did not change its name? by Martin Burger

Martin Burger — Mon, 17 Sep 2012 10:13:46 +0000

We should just lance a campaign like “Everybody is a mathematician !”

Comment on ISMP – inverse problems with uniform noise and TV does not preserve edges by Dirk

Dirk — Fri, 31 Aug 2012 19:46:11 +0000

Interesting. The “darkening” of the background shouldn’t be there (I think it is due to the finite domain on which you work, Christianes results are obtain on the whole plane and hence, the “white background” does not change color.

Comment on ISMP – inverse problems with uniform noise and TV does not preserve edges by Stefan Loock

Stefan Loock — Fri, 31 Aug 2012 15:30:32 +0000

It seems to me, that TGV behaves differently. While my TV implementation does this: http://d.pr/i/jjx0, my TGV implementation (second order, I used the primal-dual algorithm from [1]) does that: http://d.pr/i/trXt.

I took the images from your example and tried a bit until I figured out that lambda=3 gives me comparable results. Setting the same value for the TGV problem but chosing a small value (like 1/10) for the second derivative term got me those results.

It would be interesting to see, if this really holds theoretically.

[1] Antonin Chambolle and Thomas Pock. A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging. Journal of Mathematical Imaging and Vision, 40(1):120–145, 2010.

Comment on ISMP – inverse problems with uniform noise and TV does not preserve edges by Martin Benning

Martin Benning — Mon, 27 Aug 2012 15:26:51 +0000

I thought so because isotropic and anisotropic tv behave very differently in many ways, not just in terms of exact solutions of the rof model. But I might be wrong with my assumption. I made some computational tests with Christianes images and the effects seem to appear as well, however in a more anisotropic fashion, of course. But I have to check again, especially on different boundary conditions. Anyways, the more I think about it, it might not be unusual that ROF behaves like this. And it is nice to have some analytical examples on that.

Comment on ISMP – inverse problems with uniform noise and TV does not preserve edges by Dirk

Dirk — Sun, 26 Aug 2012 19:13:09 +0000

Of course, the isotropic total variation is meant here. But why do you think that the effect should not appear for isotropic variants? By the way: Are the calibrated sets for isotropic TV norms known?