Back to search
2012.07083

Hausdorff dimension estimates applied to Lagrange and Markov spectra, Zaremba theory, and limit sets of Fuchsian groups

M. Pollicott, P. Vytnova

correctmedium confidence
Category
Not specified
Journal tier
Strong Field
Processed
Sep 28, 2025, 12:55 AM

Audit review

The paper (Pollicott–Vytnova) proves the five window bounds and the global estimate dim_H(M \ L) < 0.882220 by combining Matheus–Moreira’s structural reductions with a validated transfer-operator computation, explicitly stated as Theorem 1.3 with the numerical values 0.731541, 0.855228, 0.872773, 0.812238, 0.882220 and the global bound 0.882220 . Their Section 4.1 explains the reliance on [31] for the inclusions/forbidden-word models and implements a rigorous bisection/interpolation scheme to certify dimensions of the relevant IFS limit sets . The model’s solution reproduces the same strategy: (i) symbolic coding of M and L, (ii) use of Matheus–Moreira’s gap principle and SFT reductions on specified windows, (iii) covering (M \ L) ∩ I by sums of Gauss–Cantor sets and applying dim_H(A+B) ≤ dim_H A + dim_H B, and (iv) importing the PV certified dimensions to obtain the stated bounds, hence the same final inequality. Differences are expository (sumset language versus the paper’s ‘dim(X_M)+constant’ presentation), not substantive. The numbers and logic agree with the paper’s results (see the worked parts for the intervals (√13,3.84), (3.84,3.92), and (3.92,4.01) in §4.1.1) ; the Hall–Freiman ray and the location of c_F used for the global conclusion are also recalled in the paper’s introduction .

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} strong field

\textbf{Justification:}

The manuscript presents rigorously certified upper bounds for the Hausdorff dimension of parts of the Markov–Lagrange spectra difference by combining modern validated transfer-operator techniques with the structural combinatorics of continued fractions. It improves prior rigorous bounds and confirms conjectural targets, with a method that is both efficient and broadly applicable. Exposition is generally clear; addressing a couple of small indexing/labelling slips and adding a compact mapping between windows and IFS models would further enhance readability.