2204.05008
COVER TIMES IN DYNAMICAL SYSTEMS
Natalia Jurga, Mike Todd
correcthigh confidence
- Category
- Not specified
- Journal tier
- Strong Field
- Processed
- Sep 28, 2025, 12:56 AM
- arXiv Links
- Abstract ↗PDF ↗
Audit review
The paper proves sharp two-sided bounds for the expected cover time E_mu(τ_δ) in terms of M_μ(δ) using a uniform spectral-perturbation approach for transfer operators with holes, careful control of short returns, and a Matthews-style reduction from covering to maximal hitting times; see Theorems 2.1–2.2 and the surrounding machinery in Sections 2–5 . The candidate solution sketches a different route via Abadi-type ψ-mixing hitting-time bounds for arbitrary unions of n-cylinders, a cylinder/ball sandwich, and a union-bound over a δ-grid. However, it assumes uniform exponential hitting-time estimates for all unions of n-cylinders with constants depending only on μ(A), without addressing the delicate uniform short-return phenomena that the paper treats explicitly (buffers and symbolic analysis). Under the paper’s assumptions, those uniform bounds are not automatic from ψ-mixing alone. Because these missing uniformity and short-return controls are essential to make the grid/union-bound argument rigorous in this generality, the model’s proof is incomplete/incorrect as stated, while the paper’s proof is correct.
Referee report (LaTeX)
\textbf{Recommendation:} minor revisions \textbf{Journal Tier:} strong field \textbf{Justification:} The manuscript settles the expected cover time question for a wide class of 1D dynamical systems, cleanly relating it to the minimal ball mass/Minkowski dimension. The spectral-perturbation framework and the careful handling of short returns with uniform constants are technically solid and well-motivated. Results extend to induced non-uniformly hyperbolic settings. Minor clarifications to guide readers through technical constructions would improve accessibility.