Factorization conjecture for nonstandard slow-fast vector fields

Let H(z,ε)H(z,\varepsilon) be a vector field satisfying Assumptions --, with slow manifold SεS_{\varepsilon} and linear fast fibers Nε(p)\mathcal{N}_{\varepsilon}(p) at basepoints pSεp\in S_{\varepsilon}. The tangent space to the slow manifold at pp is denoted by TpSεT_pS_{\varepsilon}. Factorization conjecture. Such vector fields may be re-factorized as

H(z,ε)=Nε(z,ε)fε(z,ε)+εG~(z,ε).H(z,\varepsilon)=N_{\varepsilon}(z,\varepsilon)f_{\varepsilon}(z,\varepsilon)+\varepsilon\widetilde{G}(z,\varepsilon).

The new objects satisfy Sε={zRn:fε(z,ε)=0}S_{\varepsilon}=\{z\in\mathbb{R}^n:f_{\varepsilon}(z,\varepsilon)=0\}; the columns of Nε(p,ε)N_{\varepsilon}(p,\varepsilon) span Nε(p)\mathcal{N}_{\varepsilon}(p); and G~(p,ε)TpSε\widetilde{G}(p,\varepsilon)\in T_pS_{\varepsilon} for every pSεp\in S_{\varepsilon}. The conjecture would provide a geometric factorization in which the slow manifold is a level set, the columns of NεN_{\varepsilon} encode the fast fibers, and the remainder is tangent to the slow manifold. The source presents this as a conjecture and gives no resolution, so its status remains open.

Sources & referencesView supporting material

Primary source

Ian Lizarraga and Martin Wechselberger, “Computational singular perturbation method for nonstandard slow-fast systems”, arXiv:1906.06049 (2019).

Progress summary

Refreshed
Open

The conjecture proposes a geometric rewrite of certain slow-fast systems, but no public proof, counterexample, or claimed solution has been found.

The conjecture asks whether, for positive ε\varepsilon, a nonstandard slow-fast vector field can be refactorized so that the slow manifold is the zero set of a new function, the columns of a new matrix describe the fast fibers, and the remainder is tangent to the slow manifold. It is explicitly presented as Conjecture 1 in the source.

Known results

  • The source discusses iterative refinements of the factorization data but gives no proof of the ε\varepsilon-dependent conjecture.
  • General existence and uniqueness of factorizations h(z)=N(z)f(z)h(z)=N(z)f(z) have only partial answers; local factorizations are available for rational vector fields, but this does not resolve the stated conjecture.

Current status (as of August 2026): The conjecture remains open; the geometric refactorization is neither proved nor disproved in the retrieved literature.

Sources

Solutions 0

No solutions have been posted yet.