English

A structured proof of Kolmogorov's Superposition Theorem

Functional Analysis 2022-08-24 v3 Machine Learning

Abstract

We present a well-structured detailed exposition of a well-known proof of the following celebrated result solving Hilbert's 13th problem on superpositions. For functions of 2 variables the statement is as follows. Kolmogorov Theorem. There are continuous functions φ1,,φ5:[0,1][0,1]\varphi_1,\ldots,\varphi_5 : [\,0, 1\,]\to [\,0,1\,] such that for any continuous function f:[0,1]2Rf: [\,0,1\,]^2\to\mathbb R there is a continuous function h:[0,3]Rh: [\,0,3\,]\to\mathbb R such that for any x,y[0,1]x,y\in [\,0, 1\,] we have f(x,y)=k=15h(φk(x)+2φk(y)).f(x,y)=\sum\limits_{k=1}^5 h\left(\varphi_k(x)+\sqrt{2}\,\varphi_k(y)\right). The proof is accessible to non-specialists, in particular, to students familiar with only basic properties of continuous functions.

Keywords

Cite

@article{arxiv.2105.00408,
  title  = {A structured proof of Kolmogorov's Superposition Theorem},
  author = {S. Dzhenzher and A. Skopenkov},
  journal= {arXiv preprint arXiv:2105.00408},
  year   = {2022}
}

Comments

English: 7+8 pages, 1+1 figures, in English and in Russian; exposition improved, Russian version added

R2 v1 2026-06-24T01:42:25.645Z