Talk:Kolmogorov–Arnold representation theorem

Recent Improvements

I posted a public domain constructive proof of this theorem by Braun and Griebel. Most of the original material is in Russian, or paywalled translations, so it was not easy to find an English language source for this. The Braun and Griebel source seems to sum up the modern understanding of this theorem (along with an accurate constructive proof), which is important since so much has changed over the last 60 years since the theorem was originally published. Modern proofs are much better and more specific than the original formulations. This should make it much easier for those who wish to learn more to find some decent source material. I also added the more specific statement of the theorem.

2nd Example

The 2nd example seems to be wrong. It has one too many nestings. 2A02:8109:AD3F:FCD4:45A7:CF44:983E:69D (talk) 20:32, 31 May 2016 (UTC)

"Nomographic function" listed at Redirects for discussion

A discussion is taking place to address the redirect Nomographic function. The discussion will occur at Wikipedia:Redirects for discussion/Log/2020 June 9#Nomographic function until a consensus is reached, and readers of this page are welcome to contribute to the discussion. 1234qwer1234qwer4 (talk) 22:31, 9 June 2020 (UTC)

First paragraph: claim that 13th problem is a corollary

Is the word "corollary" strictly correct: "It solved a more constrained form of Hilbert's thirteenth problem, so the original Hilbert's thirteenth problem is a corollary." The page for the 13th problem says: "The variant for continuous functions was resolved affirmatively in 1957 by Vladimir Arnold when he proved the Kolmogorov–Arnold representation theorem, but the variant for algebraic functions remains unresolved." Surely if it were a corollary, I'd expect it to have been resolved. Weka511 (talk) 22:17, 30 August 2024 (UTC)

Add A Fact: "Kolmogorov-Arnold networks"

I found a fact that might belong in this article. See the quote below

An April 2024 study(opens a new tab) introduced an alternative neural network design, called a Kolmogorov-Arnold network (KAN)

The fact comes from the following source:

https://www.quantamagazine.org/novel-architecture-makes-neural-networks-more-understandable-20240911/

Here is a wikitext snippet to use as a reference:

 {{Cite web |title=Novel Architecture Makes Neural Networks More Understandable |url=https://www.quantamagazine.org/novel-architecture-makes-neural-networks-more-understandable-20240911/ |website=Quanta Magazine |date=2024-09-11 |access-date=2024-10-02 |language=en |first=Steve |last=Nadis |quote=An April 2024 study(opens a new tab) introduced an alternative neural network design, called a Kolmogorov-Arnold network (KAN)}} 

This post was generated using the Add A Fact browser extension.

Lewuathe (talk) 03:53, 2 October 2024 (UTC)

The study you mention is already one of the references on the sentence "In the field of machine learning, there have been various attempts to use neural networks modeled on the Kolmogorov–Arnold representation." I don't think any further coverage is warranted at this time. Gumshoe2 (talk) 21:49, 2 October 2024 (UTC)