Abstract
We propose a fast algorithm to approximate the optimal transport distance. The main idea is to add a Fisher information regularization into the dynamical setting of the problem, originated by Benamou and Brenier. The regularized problem is shown to be smooth and strictly convex, thus many classical fast algorithms are available. In this paper, we adopt Newton’s method, which converges to the minimizer with a quadratic rate. Several numerical examples are provided.
Similar content being viewed by others
References
Benamou, J.-D., Brenier, Y.: A computational fluid mechanics solution to the Monge–Kantorovich mass transfer problem. Numerische Mathematik 84(3), 375–393 (2000)
Carlen, E.: Stochastic mechanics: a look back and a look ahead. Diffus. Quant. Theory Radic. Elem. Math. 47, 117–139 (2014)
Chambolle, A., Pock, T.: A first-order primal-dual algorithm for convex problems with applications to imaging. J. Math. Imaging Vis. 40(1), 120–145 (2011)
Chizat, L., Peyré, G., Schmitzer, B., Vialard, F.-X.: : An Interpolating Distance Between Optimal Transport and Fisher–Rao Metrics, Foundations of Computational Mathematics, pp. 1–44. Springer, Berlin (2016)
Chen, Y., Georgiou, T., Pavon, M.: On the relation between optimal transport and Schrödinger bridges: a stochastic control viewpoint. J. Optim. Theory Appl. 169(2), 671–691 (2016)
Chen, Y., Georgiou, T., Pavon, M.: Entropic and displacement interpolation: a computational approach using the Hilbert metric. SIAM J. Appl. Math. 76(6), 2375–2396 (2016)
Chow, S.-N., Huang, W., Li, Y., Zhou, H.: Fokker-Planck equations for a free energy functional or Markov process on a graph. Arch. Ration. Mech. Anal. 203(3), 969–1008 (2012)
Chow, S.-N., Li, W., Zhou, H.: Entropy dissipation of Fokker-Planck equations on graphs. arXiv:1701.04841 (2017)
Chow, S-N., Dieci, L., Li, W., Zhou, H.: Entropy dissipation semi-discretization schemes for Fokker–Planck equations. arXiv:1608.02628 (2016)
Chow, S-N., Li, W., Zhou, H.: A discrete Schrödinger equation via optimal transport on graphs. arXiv:1705.07583 (2017)
Cuturi, M.: Sinkhorn distances: Lightspeed computation of optimal transport. In: Conference on Neural Information Processing Systems (NIPS13), pp. 2292–2300, (2013)
Evans, L.: Partial differential equations and Monge–Kantorovich mass transfer. Curr. Dev. Math. 1, 65–126 (1997)
Evans, L., Gangbo, W.: Differential Equations Methods for the Monge–Kantorovich Mass Transfer Problem. American Mathematical Society, Providence, RI (1999)
Roy Frieden, B.: Science from Fisher Information: A Unification. Cambridge University Press, Cambridge (2004)
Gangbo, W., Li, W., Mou, C.: Schrödinger bridge problem on a graph via optimal transport, (2017)
Haber, E., Horesh, R.: A multilevel method for the solution of time dependent optimal transport. Numer. Math. Theory Methods Appl. 8(1), 97–111 (2015)
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)
Lee, J.D., Sun, Y., Saunders, M.A.: Proximal Newton-type methods for convex optimization. In: Advances in Neural Information Processing Systems (NIPS), vol 25 (2012)
Léonard, C.: A survey of the Schrödinger problem and some of its connections with optimal transport. arXiv preprint arXiv:1308.0215, (2013)
Maas, J.: Gradient flows of the entropy for finite Markov chains. J. Funct. Anal. 261(8), 2250–2292 (2011)
Métivier, L., Brossier, R., Mérigot, Q., Oudet, E., Virieux, J.: Measuring the misfit between seismograms using an optimal transport distance: application to full waveform inversion. Geophys. J. Int. 205(1), 345–377 (2016)
Nelson, E.: Derivation of the Schrödinger equation from Newtonian mechanics. Phys. Rev. 150(4), 1966 (1079)
Papadakis, N., Peyré, G., Oudet, E.: Optimal transport with proximal splitting. SIAM J. Imaging Sci. 7(1), 212–238 (2014). SIAM
Rubner, Y., Tomasi, C., Guibas, L.: The Earth mover’s distance as a metric for image retrieval. Int. J. Comput. Vis. 40(2), 99–121 (2000)
Rudin, L., Osher, S.: Fatemi, Emad: Nonlinear total variation based noise removal algorithms. Phys. D Nonlinear Phenom. 60, 259–268 (1992)
Santambrogio, F.: Optimal Transport for Applied Mathematicians. Birkuser, Basel (2015)
Schrödinger, E.: Quantisierung als Eigenwertproblem (zweite Mitteilung). Annalen der Physik 79, 489–527 (1926)
Villani, C.: Optimal Transport: Old and New, vol. 338. Springer Science & Business Media, Berlin (2008)
Wuchen, L.: A study of stochastic differential equations and Fokker–Planck equations with applications. Ph.D. thesis, (2016)
Wuchen, L., Ernest, R., Stanley, O., Wotao, Y., Wilfrid, G.: A fast algorithm for earth mover’s distance based on optimal transport and \(L_1\) type regularization. arXiv:1609.07092, (2016)
Wuchen, L., Penghang, Y., Stanley, O.: A fast algorithm for unbalanced \(L_1\) Monge–Katorvich problem. CAM report, (2016)
Yasue, K.: Stochastic calculus of variations. J. Funct. Anal. 41, 327–340 (1981)
Acknowledgements
We would like to thank Professors Shui-Nee Chow, Wilfrid Gangbo and Haomin Zhou for many discussions on related topics.
Author information
Authors and Affiliations
Corresponding author
Additional information
This work is partially supported by ONR Grants N000141410683, N000141210838 and DOE Grant DE-SC00183838.
Rights and permissions
About this article
Cite this article
Li, W., Yin, P. & Osher, S. Computations of Optimal Transport Distance with Fisher Information Regularization. J Sci Comput 75, 1581–1595 (2018). https://doi.org/10.1007/s10915-017-0599-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10915-017-0599-0