2104.01362
On infinitely many foliations by caustics in strictly convex non-closed billiards
Alexey Glutsyuk
correcthigh confidence
- Category
- math.DS
- Journal tier
- Specialist/Solid
- Processed
- Sep 28, 2025, 12:56 AM
- arXiv Links
- Abstract ↗PDF ↗
Audit review
The paper proves the existence of a C∞ foliation of a collar by billiard caustics for any strictly convex C∞ arc γ, and moreover, a continuum of such foliations with pairwise distinct germs along γ (Theorem 1.2 and the addendum to Theorem 1.4). The proof proceeds via the class of C∞-lifted strongly billiard-like maps, establishing a normal form with flat remainders F̃(τ,φ) = (τ + φ + flat(φ), φ + flat(φ)) (equation (1.11)) and constructing an invariant function φ̃ = φ + flat(φ), whose level sets give the invariant foliation (Theorems 1.15 and 1.18, and §2.7 for the billiard reduction) . By contrast, the candidate solution assumes a stronger, unproved C∞ symplectic conjugacy to the exact model F0(t,z) = (t + √z, z) and then pulls back the horizontal foliation {z = const}. The paper does not claim such an exact conjugacy; it works with the flat-error normal form (1.11) and an invariant first integral to produce the foliation. Thus, while the candidate’s final conclusions match the paper’s statements, the key step (global conjugacy to F0 up to the boundary) is not established in the paper and is not obviously valid in general.
Referee report (LaTeX)
\textbf{Recommendation:} minor revisions
\textbf{Journal Tier:} specialist/solid
\textbf{Justification:}
The manuscript gives a careful and convincing proof that any strictly convex C∞ arc admits a C∞ collar foliated by caustics and that there are continuum many such foliations with pairwise distinct germs. The reduction to C∞-lifted strongly billiard-like maps, the flat-error normal form, and the construction of an invariant first integral are well organized. Some expository refinements (clarifying the distinction between exact conjugacy and up-to-flat normal forms, and adding illustrations) would improve accessibility.