Luminary Research Archive
Luminary Research ArchiveIndependent scholarship
Luminary Portal

Leading bias and positive correction of heat-kernel Sinkhorn metrics

Original PDF · First-page preview

Loading original first page…

Abstract

We identify the leading bias of the population heat-kernel Sinkhorn tensor at smooth strictly positive densities on a flat torus. Its second-order quadratic-form coefficient is minus one forty-eighth of the Wasserstein Hessian of Fisher information, with a fourth-order H⁶ → L² remainder uniform under every fixed order of density-parameter differentiation. An explicit density has bias directions of both signs. A compact same-bandwidth correction removes the bias, remains positive on the full tangent complement, and preserves the original completed form space and square-root range with equivalent norms. Its sharp Hʳ → L² rates are h^(1+r/2) for 0 ≤ r ≤ 6; full L² accuracy remains first order. On immersed finite-dimensional density families, corrected metrics and intrinsic curvature have fourth-order accuracy, with relative fourth-order comparison of path distances restricted to a fixed compact parameter region. Geodesic initial-value solutions and parallel transport have the same rate on fixed time intervals within a compact parameter chart. Computational appendices give certified metric values and normalization-inclusive approximation for prescribed analytic families close to uniform, with logarithmic per-coordinate resolution under explicit coefficient and tail inputs.

Read the full paper in the original PDF.

Edition files

Available editions

  1. Edition 3 · 2026-11-02 · public
  2. Edition 2 · 2026-10-27 · public
  3. Edition 1 · 2026-10-22 · public

Research details

Subjects
Mathematics · Geometry and Optimal Transport
Keywords
Sinkhorn metrics, heat kernel, leading bias, Fisher information, positive correction, analytic densities, certified approximation
Manuscript date
2026-10-21
Reader reuse
none

Scope

We identify the leading bias of the population heat-kernel Sinkhorn tensor at smooth strictly positive densities on a flat torus.

Disclosures

Author-posted preprint. OpenAI ChatGPT and Codex assisted mathematical exploration, literature research, derivation, drafting, proof checking and reproducibility code. The author assumes responsibility for all content.

Reproduction

Complete searchable manuscript and 10-member source package with editable sources and necessary reader-facing verification material. Build instructions and computational scope are stated in the package; proofs are given in the manuscript.

References and related resources

  1. S. Schubr, Sharp convergence of heat-kernel Sinkhorn metrics, public edition 1, 2026. Luminary Research Archive.
  2. H. Lavenant, J. Luckhardt, G. Mordant, B. Schmitzer and L. Tamanini, The Riemannian geometry of Sinkhorn divergences, Ann. Inst. H. Poincaré Anal. Non Linéaire 43 (2026), no. 5, 1079--1144. doi:10.4171/AIHPC/165; arXiv:2405.04987v2.
  3. C. L. Wormell and S. Reich, Spectral convergence of diffusion maps: improved error bounds and an alternative normalisation, arXiv:2006.02037v3, 2021.
  4. R. Yao, Y.-H. Kim and G. Schiebinger, Quantitative Stability of Many-Marginal Schrödinger Bridge, arXiv:2604.15191v1, 2026.
  5. S. Schubr, First-Hit Stopping and Bias-Corrected Score Inference in Dimension Three, manuscript 2.1, public edition 2, 2026. Luminary Research Archive.
  6. L. Chizat, P. Roussillon, F. Léger, F.-X. Vialard and G. Peyré, Faster Wasserstein Distance Estimation with the Sinkhorn Divergence, Advances in Neural Information Processing Systems 33 (2020). arXiv:2006.08172v2.
  7. S. Blanes, F. Casas and A. Murua, Splitting methods for differential equations, arXiv:2401.01722v3, 2024.
  8. T. Ichinose and H. Tamura, Results on Convergence in Norm of Exponential Product Formulas and Pointwise of the Corresponding Integral Kernels, arXiv:0905.3458, 2009.
  9. S. Blanes, F. Casas, C. González and M. Thalhammer, Efficient Splitting Methods Based on Modified Potentials, Communications in Computational Physics 33 (2023), 937--961. doi:10.4208/cicp.OA-2022-0247.
  10. S. Sule, L. Evans and M. Cameron, Sharp error estimates for target measure diffusion maps with applications to the committor problem, Applied and Computational Harmonic Analysis 79 (2025), 101803. doi:10.1016/j.acha.2025.101803; arXiv:2312.14418v1.
  11. J. A. Carrillo and D. Slepčev, Example of a displacement convex functional of first order, Calculus of Variations and Partial Differential Equations 36 (2009), no. 4, 547--564. doi:10.1007/s00526-009-0243-4.
  12. G. Mordant and L. Xu, Convergence of the Sinkhorn Riemannian metric for finitely supported measures, arXiv:2609.14651v1, 2026.
  13. H. Lavenant, J. Luckhardt and B. Schmitzer, Approximation and computation of the geodesic Sinkhorn distance, arXiv:2610.02007v1, 2026.
  14. T. L. Saidi, F. Gunsilius and G. Mena. Second-order Sinkhorn Geometry and sqrt(n)-Inference. arXiv:2610.36488v1 (2026).
  15. R. G. Douglas, On majorization, factorization, and range inclusion of operators on Hilbert space, Proceedings of the American Mathematical Society 17 (1966), 413--415. doi:10.1090/S0002-9939-1966-0203464-1.
  16. N. Kessler, R. Magnet and J. Feydy, Sinkhorn Normalization of Diffusion Kernels, Proceedings of Machine Learning Research 306 (2026), 56764--56795. ICML 2026.
  17. Y. Chen, T. T. Georgiou and M. Pavon, Stochastic control liaisons: Richard Sinkhorn meets Gaspard Monge on a Schrödinger bridge, arXiv:2005.10963v2.
  18. Q. Du and J. Yang, Asymptotically Compatible Fourier Spectral Approximations of Nonlocal Allen--Cahn Equations, SIAM Journal on Numerical Analysis 54 (2016), no. 3, 1899--1919. doi:10.1137/15M1039857.
  19. Z. Shen, Z. Wang, A. Ribeiro and H. Hassani, Sinkhorn Natural Gradient for Generative Models, Advances in Neural Information Processing Systems 33 (2020). NeurIPS 2020.
  20. R. J. Berman, The Sinkhorn algorithm, parabolic optimal transport and geometric Monge--Ampère equations, Numerische Mathematik 145 (2020), 771--836. doi:10.1007/s00211-020-01127-x.
  21. L. N. Trefethen and J. A. C. Weideman, The Exponentially Convergent Trapezoidal Rule, SIAM Review 56 (2014), no. 3, 385--458. doi:10.1137/130932132.