Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
173 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Limit Theorems for Entropic Optimal Transport Maps and the Sinkhorn Divergence (2207.08683v2)

Published 18 Jul 2022 in math.ST, math.PR, and stat.TH

Abstract: We study limit theorems for entropic optimal transport (EOT) maps, dual potentials, and the Sinkhorn divergence. The key technical tool we use is a first and second-order Hadamard differentiability analysis of EOT potentials with respect to the marginal distributions, which may be of independent interest. Given the differentiability results, the functional delta method is used to obtain central limit theorems for empirical EOT potentials and maps. The second-order functional delta method is leveraged to establish the limit distribution of the empirical Sinkhorn divergence under the null. Building on the latter result, we further derive the null limit distribution of the Sinkhorn independence test statistic and characterize the correct order. Since our limit theorems follow from Hadamard differentiability of the relevant maps, as a byproduct, we also obtain bootstrap consistency and asymptotic efficiency of the empirical EOT map, potentials, and Sinkhorn divergence.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (101)
  1. Bootstrap and permutation tests of independence for point processes. The Annals of Statistics, 43(6):2537–2564, 2015.
  2. Wasserstein generative adversarial networks. In Proceedings of the 34th International Conference on Machine Learning, pages 214–223, 2017.
  3. Set-valued Analysis. Springer Science & Business Media, 2009.
  4. Gradient Flows: in Metric Spaces and in the Space of Probability Measures. Springer Science & Business Media, 2008.
  5. Asymptotics for semidiscrete entropic optimal transport. SIAM Journal on Mathematical Analysis, 54(2):1718–1741, 2022.
  6. Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration. In Proceedings of the 31st International Conference on Neural Information Processing Systems, pages 1961–1971, 2017.
  7. Central limit theorems for entropy-regularized optimal transport on finite spaces and statistical applications. Electronic Journal of Statistics, 13(2):5120–5150, 2019.
  8. Entropic optimal transport: Geometry and large deviations. Duke Mathematical Journal, 171(16):3363–3400, 2022.
  9. Resampling fewer than n𝑛nitalic_n observations: Gains, losses, and remedies for losses. Statistica Sinica, 7:1–31, 1997.
  10. Jérémie Bigot. Statistical data analysis in the Wasserstein space. ESAIM: Proceedings and Surveys, 68:1–19, 2020.
  11. Approximate bayesian computation with the Wasserstein distance. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 81(2):235–269, 2019.
  12. On parameter estimation with the Wasserstein distance. Information and Inference: A Journal of the IMA, 8(4):657–676, 2019.
  13. Quantifying distributional model risk via optimal transport. Mathematics of Operations Research, 44(2):565–600, 2019.
  14. Yann Brenier. Polar factorization and monotone rearrangement of vector-valued functions. Communications on pure and applied mathematics, 44(4):375–417, 1991.
  15. Vector quantile regression and optimal transport, from theory to numerics. Empirical Economics, pages 1–28, 2020.
  16. Vector quantile regression: an optimal transport approach. The Annals of Statistics, 44(3):1165–1192, 2016.
  17. Directional differentiability for supremum-type functionals: statistical applications. Bernoulli, 26(3):2143–2175, 2020.
  18. Convergence of entropic schemes for optimal transport and gradient flows. SIAM Journal on Mathematical Analysis, 49(2):1385–1418, 2017.
  19. Optimal transport for domain adaptation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(9):1853–1865, 2016.
  20. Monge–Kantorovich depth, quantiles, ranks and signs. The Annals of Statistics, 45(1):223–256, 2017.
  21. Inferential Wasserstein generative adversarial networks. Journal of the Royal Statistical Society: Series B (Statistical Methodology), to appear, 2021.
  22. Jackknife multiplier bootstrap: finite sample approximations to the U𝑈Uitalic_U-process supremum with applications. Probability Theory and Related Fields, 176(3):1097–1163, 2020.
  23. A differential approach to the multi-marginal Schrödinger system. SIAM Journal on Mathematical Analysis, 52(1):709–717, 2020.
  24. Wasserstein regression. Journal of the American Statistical Association, to appear, 2021.
  25. Convergence rate of general entropic optimal transport costs. arXiv preprint arXiv:2206.03347, 2022.
  26. Faster Wasserstein distance estimation with the Sinkhorn divergence. Proceedings of the 34th International Conference on Neural Information Processing Systems, 2020.
  27. A multivariate Faa di Bruno formula with applications. Transactions of the American Mathematical Society, 348(2):503–520, 1996.
  28. A formula for the time derivative of the entropic cost and applications. Journal of Functional Analysis, 280(11):108964, 2021.
  29. Marco Cuturi. Sinkhorn distances: lightspeed computation of optimal transport. In Proceedings of the 26th International Conference on Neural Information Processing Systems, pages 2292–2300, 2013.
  30. Central limit theorems for semidiscrete wasserstein distances. arXiv preprint arXiv:2202.06380, 2022.
  31. An improved central limit theorem and fast convergence rates for entropic transportation costs. arXiv preprint arXiv:2204.09105, 2022.
  32. Rates of estimation of optimal transport maps using plug-in estimators via barycentric projections. Advances in Neural Information Processing Systems, 34:29736–29753, 2021.
  33. Victor de la Peña and Evarist Giné. Decoupling: From Dependence to Independence. Springer, 1999.
  34. Richard M. Dudley. The speed of mean Glivenko-Cantelli convergence. The Annals of Mathematical Statistics, 40(1):40–50, 1969.
  35. Richard M. Dudley. Uniform Central Limit Theorems. Cambridge University Press, 1999.
  36. Quantitative stability of regularized optimal transport and convergence of Sinkhorn’s algorithm. SIAM Journal on Mathematical Analysis, 54(6):5922–5948, 2022.
  37. On the rate of convergence in Wasserstein distance of the empirical measure. Probability Theory and Related Fields, 162(3):707–738, 2015.
  38. Interpolating between optimal transport and MMD using Sinkhorn divergences. In The 22nd International Conference on Artificial Intelligence and Statistics, pages 2681–2690. PMLR, 2019.
  39. Improved training of Wasserstein GANs. In Proceedings of the International Conference on Neural Information Processing Systems (NeurIPS-2017), pages 5769–5779, 2017.
  40. Sample complexity of Sinkhorn divergences. In Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics, pages 1574–1583, 2019.
  41. Statistical inference with regularized optimal transport. arXiv preprint arXiv:2205.04283, 2022.
  42. The geometry of optimal transportation. Acta Mathematica, 177(2):113–161, 1996.
  43. Stability of entropic optimal transport and Schrödinger bridges. Journal of Functional Analysis, 283(9):109622, 2022.
  44. Multivariate ranks and quantiles using optimal transport: Consistency, rates and nonparametric testing. The Annals of Statistics, 50(2):1012–1037, 2022.
  45. Weak limits for empirical entropic optimal transport: Beyond smooth costs. arXiv preprint arXiv:2305.09745, 2023.
  46. Weak limits of entropy regularized optimal transport; potentials, plans and divergences. arXiv preprint: arXiv 2207.07427, 2022.
  47. Elliptic partial differential equations of second order, volume 224. springer, 2015.
  48. Matching for causal effects via multimarginal unbalanced optimal transport. arXiv preprint arXiv:2112.04398, 2021. updated on July 9, 2022.
  49. Distribution and quantile functions, ranks and signs in dimension d: A measure transportation approach. The Annals of Statistics, 49(2):1139 – 1165, 2021.
  50. Limit distributions and sensitivity analysis for entropic optimal transport on countable spaces. arXiv preprint arXiv:2105.00049, 2021.
  51. Asymptotics of entropy-regularized optimal transport via chaos decomposition. arXiv preprint arXiv:2011.08963, 2020.
  52. Minimax estimation of smooth optimal transport maps. The Annals of Statistics, 49(2):1166–1194, 2021.
  53. The variational formulation of the Fokker–Planck equation. SIAM journal on mathematical analysis, 29(1):1–17, 1998.
  54. Leonid V. Kantorovich. On the translocation of masses. In Doklady Akademii Nauk USSR, volume 37, pages 199–201, 1942.
  55. Empirical regularized optimal transport: Statistical theory and applications. SIAM Journal on Mathematics of Data Science, 2(2):419–443, 2020.
  56. Serge Lang. Fundamentals of differential geometry, volume 191. Springer Science & Business Media, 2012.
  57. Christian Léonard. From the schrödinger problem to the monge–kantorovich problem. Journal of Functional Analysis, 262(4):1879–1920, 2012.
  58. Christian Léonard. A survey of the Schrödinger problem and some of its connections with optimal transport. Discrete & Continuous Dynamical Systems, 34(4):1533, 2014.
  59. Entropy regularized optimal transport independence criterion. In International Conference on Artificial Intelligence and Statistics, pages 11247–11279. PMLR, 2022.
  60. A novel earth mover’s distance methodology for image matching with gaussian mixture models. In Proceedings of the IEEE International Conference on Computer Vision, pages 1689–1696, 2013.
  61. Plugin estimation of smooth optimal transport maps. arXiv preprint arXiv:2107.12364, 2021.
  62. Toshio Mikami. Monge’s problem with a quadratic cost by the zero-noise limit of h-path processes. Probability theory and related fields, 129(2):245–260, 2004.
  63. Toshio Mikami. Regularity of Schrodinger’s functional equation and mean field pdes for hℎhitalic_h-path processes. Osaka Journal of Mathematics, 56(4):831–842, 2019.
  64. Toshio Mikami. Regularity of Schrödinger’s functional equation in the weak topology and moment measures. Journal of the Mathematical Society of Japan, 73(1):99–123, 2021.
  65. Statistical bounds for entropic optimal transport: sample complexity and the central limit theorem. Advances in Neural Information Processing Systems, 32, 2019.
  66. Optimal transportation problem by stochastic optimal control. SIAM Journal on Control and Optimization, 47(3):1127–1139, 2008.
  67. Richard Nickl. On convergence and convolutions of random signed measures. Journal of Theoretical Probability, 22(1):38–56, 2009.
  68. Bracketing metric entropy rates and empirical central limit theorems for function classes of Besov-and Sobolev-type. Journal of Theoretical Probability, 20:177–199, 2007.
  69. Entropic optimal transport: Convergence of potentials. Probability Theory and Related Fields, 184(1-2):401–424, 2022.
  70. Stability of Schrödinger potentials and convergence of Sinkhorn’s algorithm. The Annals of Probability, 51(2):699–722, 2023.
  71. Estimation of Wasserstein distances in the spiked transport model. Bernoulli, to appear, 2021.
  72. Soumik Pal. On the difference between entropic cost and the optimal transport cost. arXiv preprint arXiv:1905.12206, 2019.
  73. Debiaser beware: Pitfalls of centering regularized transport maps. arXiv preprint arXiv:2202.08919, 2022.
  74. Minimax estimation of discontinuous optimal transport maps: The semi-discrete case. arXiv preprint arXiv:2301.11302, 2023.
  75. Entropic estimation of optimal transport maps. arXiv preprint arXiv:2109.12004, 2021.
  76. Large sample confidence regions based on subsamples under minimal assumptions. The Annals of Statistics, pages 2031–2050, 1994.
  77. Statistical aspects of Wasserstein distances. Annual Review of Statistics and its Application, 6:405–431, 2019.
  78. An invitation to statistics in Wasserstein space. Springer Nature, 2020.
  79. Werner Römisch. Delta method, infinite dimensional. In Encyclopedia of Statistical Sciences. Wiley, 2004.
  80. Entropic gromov-wasserstein distances: Stability, algorithms, and distributional limits. arXiv preprint arXiv:2306.00182, 2023.
  81. On the sample complexity of entropic optimal transport. arXiv e-prints, pages arXiv–2206, 2022.
  82. The earth mover’s distance as a metric for image retrieval. International Journal of Computer Vision, 40(2):99–121, 2000.
  83. Filippo Santambrogio. Optimal transport for applied mathematicians. Birkäuser, NY, 55(58-63):94, 2015.
  84. Filippo Santambrogio. {Euclidean, metric, and Wasserstein} gradient flows: An overview. Bulletin of Mathematical Sciences, 7(1):87–154, 2017.
  85. Erwin Schrödinger. Über die umkehrung der naturgesetze. Akad. Wiss. Berlin. Phys. Math., 144:144–153, 1931.
  86. Convolutional Wasserstein distances: Efficient optimal transportation on geometric domains. ACM Transactions on Graphics (TOG), 34(4):66, 2015.
  87. Limit theorems for semidiscrete optimal transport maps. arXiv preprint arXiv:2303.10155, 2023.
  88. Alexander Shapiro. On concepts of directional differentiability. Journal of Optimization Theory and Applications, 66:477–487, 1990.
  89. Nonnegative matrix factorization with earth mover’s distance metric for image analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33(8):1590–1602, 2011.
  90. Wasserstein auto-encoders. In Proceedings of the International Conference on Learning Representations (ICLR-2018), 2018.
  91. An optimal transport approach to causal inference. arXiv preprint arXiv:2108.05858, 2021.
  92. Aad W. van der Vaart. Asymptotic Statistics. Cambridge University Press, Cambridge, UK, 1998.
  93. Aad W. van der Vaart and Jon Wellner. Weak Convergence and Empirical Processes: With Applications to Statistics. Springer Science & Business Media, 1996.
  94. Cédric Villani. Topics in Optimal Transportation. Graduate Studies in Mathematics. American Mathematical Society, 2003.
  95. Cédric Villani. Optimal Transport: Old and New. Springer, 2008.
  96. J. H. van Zanten and A. W. van der Vaart. Reproducing kernel Hilbert spaces of Gaussian priors. In Pushing the limits of contemporary statistics: contributions in honor of Jayanta K. Ghosh, pages 200–222. Institute of Mathematical Statistics, 2008.
  97. Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance. Bernoulli, 25(4A):2620–2648, Nov. 2019.
  98. Wasserstein adversarial examples via projected sinkhorn iterations. In Proceedings of the 36th International Conference on Machine Learning, pages 6808–6817, 2019.
  99. Eberhard Zeidler. Applied functional analysis: main principles and their applications, volume 109. Springer Science & Business Media, 2012.
  100. Gromov-Wasserstein distances: entropic regularization, duality, and sample complexity. arXiv preprint arXiv:2212.12848, 2022.
  101. Wasserstein autoregressive models for density time series. Journal of Time Series Analysis, 43:30–52, 2022.
Citations (25)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now