Jump to content

Talk:Kolmogorov structure function

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Problems

[edit]

This article desperately needs an introduction written in plain English. — Preceding unsigned comment added by 82.36.147.173 (talk) 22:32, 3 October 2018 (UTC)[reply]

See WT:MATH#Kolmogorov structure function. Also, should this really be in the "Software" WikiProject? I see no software or direct relevance to software. — Arthur Rubin (talk) 00:12, 19 November 2011 (UTC)[reply]

I removed {{multiple issues}} presumably put there be Arthur Rubin. The article just gives the basic definitions of Kolm. Struct. Funct., adds the probability interpretation of A. Shen (Sh83 given) and gives some basic properties in VV04 that include properties of references Sh83, Vy87, and the main property proved in VV04. It is universally acknowledged that this is the main property. Not even Kolmogorov himself thought this would be the case (according to his last student L.A. Levin). In technical terms it states the distance of the Kolm Struct Funct to the diagonal (sufficiency line L in the Wikipedia article) is the randomness deficiency up to an additive logarithmic term. This means that the individual model wittnessing the Kolm Struct funct is almost the best-fitting model for the individual data in this setting. I referred to the MDL of sets as in VV04 and of probability models as VV04 and Rissanen (Ri07 given) does, and to ML since it is obvious. In my view the subject is mathematically techical but the entry is as simple as possible. It is not there to promote my interests. A previous writeup was mistakenly connected to Komogorov's work on turbulence. Since I happen to be coauthor in several papers about Kolm Struct Funct, I wrote the corect basics. So Arthur Rubin is mistaken and improperly damages the writeup if he puts the above warnings there. Paul Vitanyi

I don't understand it. It's either too technical, it's just wrong, or I'm stupid. I concede the COI issue is close, but it hasn't really been addressed. — Arthur Rubin (talk) 17:32, 28 November 2011 (UTC)[reply]
Now that it's been brought up, though, as most of the articles are yours, COI needs to be considered. I have to be careful contributing to set theory and the axiom of choice, because my parents' books are among the most used books in the field. — Arthur Rubin (talk) 17:35, 28 November 2011 (UTC)[reply]

Contemporary Definition

[edit]

What is meant by "change into " - a string into a set of strings - which contains one string again, just "setified"? Is there a kind of underlying encoding ongoing? I would appreciate some clarification and a reference. — Preceding unsigned comment added by 2A02:2788:925:E21B:451F:6935:4F29:9AD7 (talk) 21:07, 16 February 2021 (UTC)[reply]