Quadratic forms between spheres and the non-existence of sums of squares formulae (Q4720404)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3992436
Language Label Description Also known as
default for all languages
No label defined
    English
    Quadratic forms between spheres and the non-existence of sums of squares formulae
    scientific article; zbMATH DE number 3992436

      Statements

      Quadratic forms between spheres and the non-existence of sums of squares formulae (English)
      0 references
      1986
      0 references
      bilinear forms
      0 references
      Hopf construction
      0 references
      sums of squares formulas
      0 references
      0 references
      For integers r,s,n define an [r,s,n] formula to be one of the type \((x^ 2_ 1+x^ 2_ 2+...+x^ 2_ r)(y^ 2_ 1+y^ 2_ 2+...+y^ 2_ s)=f^ 2_ 1+f^ 2_ 2+...+f^ 2_ n,\) where \(f_ 1,f_ 2,...,f_ n\) are bilinear forms with real coefficients in \(x_ 1,...,x_ r\) and \(y_ 1,...,y_ s.\) In 1898 Hurwitz asked: which r,s,n are possible ? Such a formula is equivalent to a bilinear map \(f: {\mathbb{R}}^ r\times {\mathbb{R}}^ s\to {\mathbb{R}}^ n\) satisfying the norm condition: \(| f(x,y)| =| x| | y|\), for all \(x\in {\mathbb{R}}^ r\), \(y\in {\mathbb{R}}^ s\). Such a map f can be studied using its Hopf construction, \(h(f): S^{r+s-i}\to S^ n,\) a map between spheres. Let \(F: S^ m\to S^ n\) be an arbitrary homogeneous quadratic map between spheres. For a point q in the image of F, the author proves that the inverse image \(F^{-1}(q)\) is a great sphere in \(S^ m\). If the dimension of that great sphere is k-1 there is an associated non-singular bilinear map \({\mathbb{R}}^ k\times {\mathbb{R}}^{m+1-k}\to {\mathbb{R}}^ n.\) The Hopf construction of this ''hidden'' bilinear map is homotopic to the original map F. Furthermore this hidden map is homotopic to another normed bilinear map. Then behind any [r,s,n] formula there are other ''hidden'' sums of squares formulas of types \([k,r+s-k,n]\). The author uses these geometric ideas and some non-trivial homotopy theory to deduce that certain sums of squares formulas cannot exist, notably those of types [12,12,20] and [16,16,24]. He also deduces that there is no non-constant quadratic form \(S^{25}\to S^{23}\). In ''Sums of squares formulae with integer coefficients'', Can. Math. Bull. (to appear; Zbl 0586.10013) the author proves that [16,16,28] formulas with integer coefficients cannot exist. Recently the author and K. Y. Lam have proved this result for real coefficients as well.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references