Back to search
2402.08425

Transfer Operators from Batches of Unpaired Points via Entropic Transport Kernels

Florian Beier, Hancheng Bi, Clément Sarrazin, Bernhard Schmitzer, Gabriele Steidl

correctmedium confidence
Category
math.DS
Journal tier
Strong Field
Processed
Sep 28, 2025, 12:56 AM

Audit review

The paper’s Corollary 13 states exactly the Γ-convergence and minimizer stability result under compactness of Q, uniform positivity, and Γ-convergence of the indicators, and proves it by combining joint almost-sure convergence J_N^M(q_N)→J_M(q) along uniformly convergent sequences (Corollary 12) with the Γ-convergence of the indicator constraints and the fundamental theorem of Γ-convergence . The candidate solution follows the same structure, adding a Lipschitz argument and a countable-dense SLLN diagonalization to obtain an almost-sure event uniform over Q; the liminf/limsup steps and the minimizer conclusions match the paper’s proof. No substantive gaps relative to the paper’s assumptions were found.

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} strong field

\textbf{Justification:}

The result is correct and the proof is clear and standard for the setting considered. Minor clarifications could further improve readability (e.g., explicitly stating the topology in all Γ-convergence assertions and recalling the positivity assumption where the log continuity is used). The contribution fits well within the emerging literature on learning transfer operators and statistical consistency of kernelized estimators.