2010.11654
TIME-RESTRICTED SENSITIVITY AND ENTROPY
Kairan Liu, Leiye Xu, Ruifeng Zhang
incompletehigh confidence
- Category
- math.DS
- Journal tier
- Strong Field
- Processed
- Sep 28, 2025, 12:55 AM
- arXiv Links
- Abstract ↗PDF ↗
Audit review
The paper’s Theorem A states a_μ(x) = 1/h_μ(T,x) for μ-a.e. x and, in the ergodic case, T is μ-restricted sensitive iff h_μ(T)>0, aligning with the candidate’s claim. The paper’s route: define a_μ(x,δ) via first sensitive times, show 1/a_μ(x) = lim_{δ→0} inf_n[-log μ(B_n(x,δ))]/n (Lemma 3.5), and identify this limit with Brin–Katok local entropy (Section 2.2) to conclude Theorem A (proof immediately after Lemma 3.5). However, the proof of Lemma 3.4 contains a gap: it implicitly treats lim_{δ→0} μ(B_n(x,δ)) as μ({x}) for n≥1 (the line transitioning to “min_{1≤n≤N0} −log μ({x})/n”); in general B_n(x,δ)→∩_{i=0}^n T^{-i}({T^i x}) as δ→0, not {x}, unless injectivity or a similar property holds. This step (used to force μ({x})>0 and obtain a contradiction) needs repair; a corrected argument can bound the min by the n=0 term to reach the same contradiction but is not written as such in the text (see Lemma 3.4’s proof steps in the PDF). Aside from this fixable issue, the intended argument matches standard techniques and the main result is correct. The candidate solution is correct and gives a direct two-sided estimate using Brin–Katok: (i) for small ε, use liminf to get μ(B_n(x,ε)) ≤ e^{-(h(x)-η)n} for all large n and compare to μ(V) to get the upper bound a_μ(x) ≤ 1/h(x), and (ii) use limsup to produce Bowen balls of relatively large measure to block faster rates, giving a_μ(x) ≥ 1/h(x). The candidate should explicitly note that μ(V\B_n(x,ε)) ≥ μ(V)−μ(B_n(x,ε)) > 0 to match the paper’s positive-measure quantifier in the definition of restricted sensitivity. Overall: the theorem is right; the paper’s proof has a minor but real gap in Lemma 3.4; the model’s proof is sound when quantifier/measure details are made explicit. Citations: Theorem A statement is in the introduction (Theorem A) and proved after Lemma 3.5; key definitions and lemmas appear in Def. 3.1, Lemmas 3.2–3.5, with the Brin–Katok local entropy formula summarized in §2.2 (e.g., “for μ-a.e. x … the limsup and liminf agree”). See the arXiv PDF excerpts for these items (Theorem A and its proof, Lemma 3.5, Lemma 3.4 and its contentious step, and the definition of measure-theoretic restricted sensitivity). . Brin–Katok is summarized in §2.2 of the paper.
Referee report (LaTeX)
\textbf{Recommendation:} minor revisions
\textbf{Journal Tier:} strong field
\textbf{Justification:}
The paper delivers a crisp and useful identity between a time-restricted sensitivity rate and Brin–Katok local entropy, with a natural ergodic corollary. The definitions are interesting and the overall structure is sound. However, the proof of Lemma 3.4 contains a nontrivial but easily correctable limit step (passing from Bowen balls to an atom at x for n≥1). Adding the corrected argument would remove this gap. With that fix, the paper merits publication.