Sampling conditioned hypoelliptic diffusions. (English) Zbl 1219.60062

The authors consider the problem of sampling conditioned paths of the second-order SDE of the form \(m\ddot{x}(t)=f(x(t))-\dot{x}(t)+\dot{w}(t)\) conditioned on \(x(0)=x_-\) and \(x(T)=x_+\). This equation describes, for example, the time evolution of a noisy mechanical system with inertia and friction. Rewriting this equation as a system of first-order SDEs for \(x\) and \(\dot{x}\) leads to a drift which is in general not a gradient and, since the noise only acts on \(\dot{x}\), one obtain a singular diffusion matrix. To solve the problem, a fourth-order parabolic type SPDE with nonlinear drift is derived so that its boundary conditions and the drift term supply that the conditioned distribution of the initial equation is stationary for this SPDE. The nonlinear drift has very weak dissipativity and regularity properties that creates additional technical difficulties.


60H15 Stochastic partial differential equations (aspects of stochastic analysis)
60J60 Diffusion processes
37H10 Generation, random and stochastic difference and differential equations
Full Text: DOI arXiv


[1] Apte, A., Hairer, M., Stuart, A. M. and Voss, J. (2007). Sampling the posterior: An approach to non-Gaussian data assimilation. Phys. D 230 50-64. · Zbl 1113.62026
[2] Blömker, D., Maier-Paape, S. and Wanner, T. (2001). Spinodal decomposition for the Cahn-Hilliard-Cook equation. Comm. Math. Phys. 223 553-582. · Zbl 0993.60061
[3] Da Prato, G. and Debussche, A. (1996). Stochastic Cahn-Hilliard equation. Nonlinear Anal. 26 241-263. · Zbl 0838.60056
[4] Da Prato, G. and Zabczyk, J. (1992). Stochastic Equations in Infinite Dimensions. Encyclopedia of Mathematics and Its Applications 44 . Cambridge Univ. Press, Cambridge. · Zbl 0761.60052
[5] Hairer, M. (2009). An introduction to stochastic PDEs. Available at . · Zbl 1186.60052
[6] Hairer, M. (2010). Singular perturbations to semilinear stochastic heat equations. Preprint. Available at . · Zbl 1251.60052
[7] Hairer, M., Stuart, A. M. and Voss, J. (2007). Analysis of SPDEs arising in path sampling. II. The nonlinear case. Ann. Appl. Probab. 17 1657-1706. · Zbl 1140.60329
[8] Hairer, M., Stuart, A. M., Voss, J. and Wiberg, P. (2005). Analysis of SPDEs arising in path sampling. I. The Gaussian case. Commun. Math. Sci. 3 587-603. · Zbl 1138.60326
[9] Hairer, M., Stuart, A. and Voß, J. (2009). Sampling conditioned diffusions. In Trends in Stochastic Analysis. London Mathematical Society Lecture Note Series 353 159-185. Cambridge Univ. Press, Cambridge. · Zbl 1203.60100
[10] Kato, T. (1966). Perturbation Theory for Linear Operators . Springer, Berlin. · Zbl 0148.12601
[11] Lauritsen, K. B., Cuerno, R. and Makse, H. A. (1996). Noisy Kuramoto-Sivashinsky equation for an erosion model. Phys. Rev. E 54 3577-3580.
[12] Röckner, M. and Ma, Z. M. (1992). Introduction to the Theory of ( nonsymmetric ) Dirichlet Forms . Springer, Berlin.
[13] Reed, M. and Simon, B. (1972). Methods of Modern Mathematical Physics. I. Functional Analysis . Academic Press, New York. · Zbl 0459.46001
[14] Reznikoff, M. G. and Vanden-Eijnden, E. (2005). Invariant measures of stochastic partial differential equations and conditioned diffusions. C. R. Math. Acad. Sci. Paris 340 305-308. · Zbl 1063.60092
[15] Stuart, A. M., Voss, J. and Wiberg, P. (2004). Conditional path sampling of SDEs and the Langevin MCMC method. Commun. Math. Sci. 2 685-697. · Zbl 1082.65004
[16] Triebel, H. (1983). Theory of Function Spaces. Monographs in Mathematics 78 . Birkhäuser, Basel. · Zbl 0546.46027
[17] Triebel, H. (1992). Theory of Function Spaces. II. Monographs in Mathematics 84 . Birkhäuser, Basel. · Zbl 0763.46025
[18] Triebel, H. (2006). Theory of Function Spaces. III. Monographs in Mathematics 100 . Birkhäuser, Basel. · Zbl 1104.46001
[19] Zabczyk, J. (1989). Symmetric solutions of semilinear stochastic equations. In Stochastic Partial Differential Equations and Applications , II ( Trento , 1988). Lecture Notes in Math. 1390 237-256. Springer, Berlin. · Zbl 0701.60060
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.