Abstract
We present the fundamentals of a measure transport approach to sampling. The idea is to construct a deterministic coupling – i.e., a transport map – between a complex “target” probability measure of interest and a simpler reference measure. Given a transport map, one can generate arbitrarily many independent and unweighted samples from the target simply by pushing forward reference samples through the map. If the map is endowed with a triangular structure, one can also easily generate samples from conditionals of the target measure. We consider two different and complementary scenarios: first, when only evaluations of the unnormalized target density are available and, second, when the target distribution is known only through a finite collection of samples. We show that in both settings, the desired transports can be characterized as the solutions of variational problems. We then address practical issues associated with the optimization-based construction of transports: choosing finite-dimensional parameterizations of the map, enforcing monotonicity, quantifying the error of approximate transports, and refining approximate transports by enriching the corresponding approximation spaces. Approximate transports can also be used to “Gaussianize” complex distributions and thus precondition conventional asymptotically exact sampling schemes. We place the measure transport approach in broader context, describing connections with other optimization-based samplers, with inference and density estimation schemes using optimal transport, and with alternative transformation-based approaches to simulation. We also sketch current work aimed at the construction of transport maps in high dimensions, exploiting essential features of the target distribution (e.g., conditional independence, low-rank structure). The approaches and algorithms presented here have direct applications to Bayesian computation and to broader problems of stochastic simulation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
See [61] for a discussion on the asymptotic equivalence of the K–L divergence and Hellinger distance in the context of transport maps.
- 2.
The lexicographic order on \( \mathbb{R}^n \) is defined as follows. For \(x,y \in \mathbb{R}^{n}\), we define x⪯y if and only if either x = y or the first nonzero coordinate in y − x is positive [32]. ⪯ is a total order on \( \mathbb{R}^n \). Thus, we define T to be a monotone increasing function if and only if x⪯y implies T(x)⪯T(y). Notice that monotonicity can be defined with respect to any order on \( \mathbb{R}^n \) (e.g., ⪯ need not be the lexicographic order). There is no natural order on \( \mathbb{R}^n \) except when n = 1. It is easy to verify that for a triangular function T, monotonicity with respect to the lexicographic order is equivalent to the following: the kth component of T is a monotone function of the kth input variable.
- 3.
Roots can be found using, for instance, Newton’s method. When a component of the inverse transport is parameterized using polynomials, however, then a more robust root-finding approach is to use a bisection method based on Sturm sequences (e.g., [63]).
References
Adams, M.R., Guillemin, V.: Measure Theory and Probability. Birkhäuser Basel (1996)
Ambrosio, L., Gigli, N.: A user’s guide to optimal transport. In: Benedetto, P., Michel, R. (eds) Modelling and Optimisation of Flows on Networks, pp. 1–155. Springer, Berlin/Heidelberg (2013)
Andrieu, C., Moulines, E.: On the ergodicity properties of some adaptive MCMC algorithms. Ann. Appl. Probab. 16(3), 1462–1505 (2006)
Angenent, S., Haker, S., Tannenbaum, A.: Minimizing flows for the Monge–Kantorovich problem. SIAM J. Math. Anal. 35(1), 61–97 (2003)
Atkins, E., Morzfeld, M., Chorin, A.J.: Implicit particle methods and their connection with variational data assimilation. Mon. Weather Rev. 141(6), 1786–1803 (2013)
Attias, H.: Inferring parameters and structure of latent variable models by variational Bayes. In: Proceedings of the Fifteenth Conference on Uncertainty in Artificial Intelligence, Stockholm, pp. 21–30. Morgan Kaufmann Publishers Inc. (1999)
Bangerth, W., Rannacher, R.: Adaptive Finite Element Methods for Differential Equations. Birkhäuser Basel (2013)
Bardsley, J.M., Solonen, A., Haario, H., Laine, M.: Randomize-then-optimize: a method for sampling from posterior distributions in nonlinear inverse problems. SIAM J. Sci. Comput. 36(4), A1895–A1910 (2014)
Beaumont, M.A., Zhang, W., Balding, D.J.: Approximate Bayesian computation in population genetics. Genetics 162(4), 2025–2035 (2002)
Benamou, J.D., Brenier, Y.: A computational fluid mechanics solution to the Monge-Kantorovich mass transfer problem. Numer. Math. 84(3), 375–393 (2000)
Bernard, P., Buffoni, B.: Optimal mass transportation and Mather theory. J. Eur. Math. Soc. 9, 85–121 (2007)
Bigoni, D., Spantini, A., Marzouk, Y.: On the computation of monotone transports (2016, preprint)
Bonnotte, N.: From Knothe’s rearrangement to Brenier’s optimal transport map. SIAM J. Math. Anal. 45(1), 64–87 (2013)
Box, G., Cox, D.: An analysis of transformations. J. R. Stat. Soc. Ser. B 26(2), 211–252 (1964)
Brenier, Y.: Polar factorization and monotone rearrangement of vector-valued functions. Commun. Pure Appl. Math. 44(4), 375–417 (1991)
Brooks, S., Gelman, A., Jones, G., Meng, X.L. (eds.): Handbook of Markov Chain Monte Carlo. Boca Raton (2011)
Calderhead, B.: A general construction for parallelizing Metropolis-Hastings algorithms. Proc. Natl. Acad. Sci. 111(49), 17408–17413 (2014)
Carlier, G., Galichon, A., Santambrogio, F.: From Knothe’s transport to Brenier’s map and a continuation method for optimal transport. SIAM J. Math. Anal. 41(6), 2554–2576 (2010)
Champion, T., De Pascale, L.: The Monge problem in \(\mathbb{R}^{d}\). Duke Math. J. 157(3), 551–572 (2011)
Chib, S., Jeliazkov, I.: Marginal likelihood from the Metropolis-Hastings output. J. Am. Stat. Assoc. 96(453), 270–281 (2001)
Chorin, A., Morzfeld, M., Tu, X.: Implicit particle filters for data assimilation. Commun. Appl. Math. Comput. Sci. 5(2), 221–240 (2010)
Chorin, A.J., Tu, X.: Implicit sampling for particle filters. Proc. Natl. Acad. Sci. 106(41), 17,249–17,254 (2009)
Csilléry, K., Blum, M.G.B., Gaggiotti, O.E., François, O.: Approximate Bayesian computation (ABC) in practice. Trends Ecol. Evol. 25(7), 410–8 (2010)
Cui, T., Law, K.J.H., Marzouk, Y.M.: Dimension-independent likelihood-informed MCMC. J. Comput. Phys. 304(1), 109–137 (2016)
Cui, T., Martin, J., Marzouk, Y.M., Solonen, A., Spantini, A.: Likelihood-informed dimension reduction for nonlinear inverse problems. Inverse Probl. 30(11), 114,015 (2014)
Del Moral, P., Doucet, A., Jasra, A.: Sequential Monte Carlo samplers. J. R. Stat. Soc. B 68(3), 411–436 (2006)
Feyel, D., Üstünel, A.S.: Monge-Kantorovitch measure transportation and Monge-Ampere equation on Wiener space. Probab. Theory Relat. Fields 128(3), 347–385 (2004)
Fox, C.W., Roberts, S.J.: A tutorial on variational Bayesian inference. Artif. Intell. Rev. 38(2), 85–95 (2012)
Gautschi, W.: Orthogonal polynomials: applications and computation. Acta Numer. 5, 45–119 (1996)
Gelman, A., Carlin, J.B., Stern, H.S., Rubin, D.B.: Bayesian Data Analysis, 2nd edn. Chapman and Hall, Boca Raton (2003)
Gelman, A., Meng, X.L.: Simulating normalizing constants: from importance sampling to bridge sampling to path sampling. Stat. Sci. 13, 163–185 (1998)
Ghorpade, S., Limaye, B.V.: A Course in Multivariable Calculus and Analysis. Springer, New York (2010)
Gilks, W., Richardson, S., Spiegelhalter, D. (eds.): Markov Chain Monte Carlo in Practice. Chapman and Hall, London (1996)
Girolami, M., Calderhead, B.: Riemann manifold Langevin and Hamiltonian Monte Carlo methods. J. R. Stat. Soc. Ser. B 73, 1–37 (2011)
Goodman, J., Lin, K.K., Morzfeld, M.: Small-noise analysis and symmetrization of implicit Monte Carlo samplers. Commun. Pure Appl. Math. 2–4, n/a (2015)
Gorham, J., Mackey, L.: Measuring sample quality with Stein’s method. In: Advances in Neural Information Processing Systems, Montréal, Canada, pp. 226–234 (2015)
Haario, H., Saksman, E., Tamminen, J.: An adaptive metropolis algorithm. Bernoulli 7(2), 223–242 (2001)
Haber, E., Rehman, T., Tannenbaum, A.: An efficient numerical method for the solution of the L 2 optimal mass transfer problem. SIAM J. Sci. Comput. 32(1), 197–211 (2010)
Huan, X., Parno, M., Marzouk, Y.: Adaptive transport maps for sequential Bayesian optimal experimental design (2016, preprint)
Jaakkola, T.S., Jordan, M.I.: Bayesian parameter estimation via variational methods. Stat. Comput. 10(1), 25–37 (2000)
Kim, S., Ma, R., Mesa, D., Coleman, T.P.: Efficient Bayesian inference methods via convex optimization and optimal transport. IEEE Symp. Inf. Theory 6, 2259–2263 (2013)
Kleywegt, A., Shapiro, A., Homem-de-Mello, T.: The sample average approximation method for stochastic discrete optimization. SIAM J. Optim. 12(2), 479–502 (2002)
Kushner, H., Yin, G.: Stochastic Approximation and Recursive Algorithms and Applications. Springer, New York (2003)
Laparra, V., Camps-Valls, G., Malo, J.: Iterative gaussianization: from ICA to random rotations. IEEE Trans. Neural Netw. 22(4), 1–13 (2011)
Laurence, P., Pignol, R.J., Tabak, E.G.: Constrained density estimation. In: Quantitative Energy Finance, pp. 259–284. Springer, New York (2014)
Le Maitre, O., Knio, O.M.: Spectral Methods for Uncertainty Quantification: With Applications to Computational Fluid Dynamics. Springer, Dordrecht/New York (2010)
Litvinenko, A., Matthies, H.G.: Inverse Problems and Uncertainty Quantification. arXiv:1312.5048 (2013)
Litvinenko, A., Matthies, H.G.: Uncertainty quantification and non-linear Bayesian update of PCE coefficients. PAMM 13(1), 379–380 (2013)
Liu, J.S.: Monte Carlo Strategies in Scientific Computing. Springer, New York (2004)
Loeper, G., Rapetti, F.: Numerical solution of the Monge–Ampère equation by a Newton’s algorithm. Comptes Rendus Math. 340(4), 319–324 (2005)
Luenberger, D.G.: Optimization by Vector Space Methods. Wiley, New York (1968)
Marin, J.M., Pudlo, P., Robert, C.P., Ryder, R.J.: Approximate Bayesian computational methods. Stat. Comput. 22(6), 1167–1180 (2012)
Martin, J., Wilcox, L., Burstedde, C., Ghattas, O.: A stochastic Newton MCMC method for large-scale statistical inverse problems with application to seismic inversion. SIAM J. Sci. Comput. 34(3), 1460–1487 (2012)
Matthies, H.G., Zander, E., Rosić, B.V., Litvinenko, A., Pajonk, O.: Inverse problems in a Bayesian setting. arXiv:1511.00524 (2015)
McCann, R.: Existence and uniqueness of monotone measure-preserving maps. Duke Math. J. 80(2), 309–323 (1995)
Meng, X.L., Schilling, S.: Warp bridge sampling. J. Comput. Graph. Stat. 11(3), 552–586 (2002)
Monge, G.: Mémoire sur la théorie des déblais et de remblais. In: Histoire de l’Académie Royale des Sciences de Paris, avec les Mémoires de Mathématique et de Physique pour la même année, pp. 666–704 (1781)
Morzfeld, M., Chorin, A.J.: Implicit particle filtering for models with partial noise, and an application to geomagnetic data assimilation. arXiv:1109.3664 (2011)
Morzfeld, M., Tu, X., Atkins, E., Chorin, A.J.: A random map implementation of implicit filters. J. Comput. Phys. 231(4), 2049–2066 (2012)
Morzfeld, M., Tu, X., Wilkening, J., Chorin, A.: Parameter estimation by implicit sampling. Commun. Appl. Math. Comput. Sci. 10(2), 205–225 (2015)
Moselhy, T., Marzouk, Y.: Bayesian inference with optimal maps. J. Comput. Phys. 231(23), 7815–7850 (2012)
Neal, R.M.: MCMC using Hamiltonian dynamics. In: Brooks, S., Gelman, A., Jones, G.L., Meng, X.L. (eds.) Handbook of Markov Chain Monte Carlo, chap. 5, pp. 113–162. Taylor and Francis, Boca Raton (2011)
Parno, M.: Transport maps for accelerated Bayesian computation. Ph.D. thesis, Massachusetts Institute of Technology (2014)
Parno, M., Marzouk, Y.: Transport Map Accelerated Markov Chain Monte Carlo. arXiv:1412.5492 (2014)
Parno, M., Moselhy, T., Marzouk, Y.: A Multiscale Strategy for Bayesian Inference Using Transport Maps. arXiv:1507.07024 (2015)
Ramsay, J.: Estimating smooth monotone functions. J. R. Stat. Soc. Ser. B 60(2), 365–375 (1998)
Reich, S.: A nonparametric ensemble transform method for Bayesian inference. SIAM J. Sci. Comput. 35(4), A2013–A2024 (2013)
Renegar, J.: A Mathematical View of Interior-Point Methods in Convex Optimization, vol. 3. SIAM, Philadelphia (2001)
Robert, C.P., Casella, G.: Monte Carlo Statistical Methods, 2nd edn. Springer, New York (2004)
Rosenblatt, M.: Remarks on a multivariate transformation. Ann. Math. Stat. 23(3), 470–472 (1952)
Rosić, B.V., Litvinenko, A., Pajonk, O., Matthies, H.G.: Sampling-free linear Bayesian update of polynomial chaos representations. J. Comput. Phys. 231(17), 5761–5787 (2012)
Saad, G., Ghanem, R.: Characterization of reservoir simulation models using a polynomial chaos-based ensemble Kalman filter. Water Resour. Res. 45(4), n/a (2009)
Smith, A., Doucet, A., de Freitas, N., Gordon, N. (eds.): Sequential Monte Carlo Methods in Practice. Springer, New York (2001)
Spall, J.C.: Introduction to Stochastic Search and Optimization: Estimation, Simulation, and Control, vol. 65. Wiley, Hoboken (2005)
Spantini, A., Marzouk, Y.: On the low-dimensional structure of measure transports (2016, preprint)
Spantini, A., Solonen, A., Cui, T., Martin, J., Tenorio, L., Marzouk, Y.: Optimal low-rank approximations of Bayesian linear inverse problems. SIAM J. Sci. Comput. 37(6), A2451–A2487 (2015)
Stavropoulou, F., Müller, J.: Parameterization of random vectors in polynomial chaos expansions via optimal transportation. SIAM J. Sci. Comput. 37(6), A2535–A2557 (2015)
Strang, G., Fix, G.J.: An Analysis of the Finite Element Method, vol. 212. Prentice-Hall, Englewood Cliffs (1973)
Stuart, A.M.: Inverse problems: a Bayesian perspective. Acta Numer. 19, 451–559 (2010)
Sullivan, A.B., Snyder, D.M., Rounds, S.A.: Controls on biochemical oxygen demand in the upper Klamath River, Oregon. Chem. Geol. 269(1-2), 12–21 (2010)
Tabak, E., Turner, C.V.: A family of nonparametric density estimation algorithms. Communications on Pure and Applied Mathematics 66(2), 145–164 (2013)
Tabak, E.G., Trigila, G.: Data-driven optimal transport. Commun. Pure Appl. Math. 10, 1002 (2014)
Thode, H.C.: Testing for Normality, vol. 164. Marcel Dekker, New York (2002)
Villani, C.: Topics in Optimal Transportation, vol. 58. American Mathematical Society, Providence (2003)
Villani, C.: Optimal Transport: Old and New, vol. 338. Springer, Berlin/Heidelberg (2008)
Wackernagel, H.: Multivariate Geostatistics: An Introduction with Applications. Springer-Verlag Berlin Heidelberg (2013)
Wainwright, M.J., Jordan, M.I.: Graphical models, exponential families, and variational inference. Found. Trends Mach. Learn. 1(1–2), 1–305 (2008)
Wang, L.: Methods in Monte Carlo computation, astrophysical data analysis and hypothesis testing with multiply-imputed data. Ph.D. thesis, Harvard University (2015)
Wilkinson, D.J.: Stochastic Modelling for Systems Biology. CRC Press, Boca Raton (2011)
Wright, S.J., Nocedal, J.: Numerical Optimization, vol. 2. Springer, New York (1999)
Xiu, D., Karniadakis, G.: The Wiener-Askey polynomial chaos for stochastic differential equations. SIAM J. Sci. Comput. 24(2), 619–644 (2002)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing Switzerland
About this entry
Cite this entry
Marzouk, Y., Moselhy, T., Parno, M., Spantini, A. (2017). Sampling via Measure Transport: An Introduction. In: Ghanem, R., Higdon, D., Owhadi, H. (eds) Handbook of Uncertainty Quantification. Springer, Cham. https://doi.org/10.1007/978-3-319-12385-1_23
Download citation
DOI: https://doi.org/10.1007/978-3-319-12385-1_23
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-12384-4
Online ISBN: 978-3-319-12385-1
eBook Packages: Mathematics and StatisticsReference Module Computer Science and Engineering