Back to search
2505.00276

Topological State Space Inference for Dynamical Systems

Mishal Assif P K, Yuliy Baryshnikov

correctmedium confidence
Category
Not specified
Journal tier
Specialist/Solid
Processed
Sep 28, 2025, 12:56 AM

Audit review

The paper defines slack distance (Definition 3.1) and proves Lemma 3.2: composing two certificates (s, ε) and (s′, ε′) yields a certificate for the endpoints with parameters at most (s+s′, ε+ε′). Its proof is concise, using a union-of-bad-indices argument, and then states as a corollary that any convex, 1-homogeneous aggregator d_π yields a triangle inequality and a Vietoris–Rips bifiltration in (t, r) (monotonicity) . The candidate solution provides a more explicit index-intersection construction and a trimming step to achieve exactly s+s′ in the nondegenerate regime, plus a detailed subadditivity argument for d_π and a clear monotonicity proof for the bifiltration. The two arguments agree in substance; the model’s proof supplies missing details and edge-case handling that the paper sketches but does not fully spell out. Overall, both are correct, with different levels of rigor.

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} specialist/solid

\textbf{Justification:}

The work presents a practical slack-distance framework and a two-parameter Vietoris–Rips construction for topological inference from dynamical observations. The central lemma and corollary are correct and align with the computational pipeline. Exposition would benefit from fuller proofs (especially index alignment and edge cases) and a more explicit justification of the triangle inequality from convex 1-homogeneity. These are minor presentational improvements rather than substantive gaps.