## Swipe to navigate through the chapters of this book

Published in:

2020 | OriginalPaper | Chapter

# Learning Stochastic Dynamical Systems via Bridge Sampling

Authors : Harish S. Bhat, Shagun Rawat

Published in:

Publisher:

## Abstract

We develop algorithms to automate discovery of stochastic dynamical system models from noisy, vector-valued time series. By discovery, we mean learning both a nonlinear drift vector field and a diagonal diffusion matrix for an Itô stochastic differential equation in $$\mathbb {R}^d$$. We parameterize the vector field using tensor products of Hermite polynomials, enabling the model to capture highly nonlinear and/or coupled dynamics. We solve the resulting estimation problem using expectation maximization (EM). This involves two steps. We augment the data via diffusion bridge sampling, with the goal of producing time series observed at a higher frequency than the original data. With this augmented data, the resulting expected log likelihood maximization problem reduces to a least squares problem. We provide an open-source implementation of this algorithm. Through experiments on systems with dimensions one through eight, we show that this EM approach enables accurate estimation for multiple time series with possibly irregular observation times. We study how the EM method performs as a function of the amount of data augmentation, as well as the volume and noisiness of the data.
Literature
1.
Archambeau, C., Opper, M., Shen, Y., Cornford, D., Shawe-Taylor, J.S.: Variational inference for diffusion processes. In: Advances in Neural Information Processing Systems, pp. 17–24 (2008)
2.
Batz, P., Ruttor, A., Opper, M.: Variational estimation of the drift for stochastic differential equations from the empirical density. J. Stat. Mech: Theory Exp. 2016(8), 083404 (2016)
3.
Batz, P., Ruttor, A., Opper, M.: Approximate Bayes learning of stochastic differential equations. Phys. Rev. E 98, 022109 (2018)
4.
Bhat, H.S., Madushani, R.W.M.A.: Nonparametric adjoint-based inference for stochastic differential equations. In: 2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA), pp. 798–807 (2016)
5.
Bhattacharya, R.N., Waymire, E.C.: Stochastic Processes with Applications. SIAM (2009)
6.
Brunton, S.L., Proctor, J.L., Kutz, J.N.: Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc. Nat. Acad. Sci. 113(15), 3932–3937 (2016)
7.
Chen, S., Shojaie, A., Witten, D.M.: Network reconstruction from high-dimensional ordinary differential equations. J. Am. Stat. Assoc. 112(520), 1697–1707 (2017)
8.
Ghahramani, Z., Roweis, S.T.: Learning nonlinear dynamical systems using an EM algorithm. In: Advances in Neural Information Processing Systems (NIPS), pp. 431–437 (1999)
9.
Kloeden, P.E., Platen, E.: Numerical Solution of Stochastic Differential Equations. Springer, Heidelberg (2011). https://​doi.​org/​10.​1007/​978-3-662-12616-5
10.
Mangan, N.M., Brunton, S.L., Proctor, J.L., Kutz, J.N.: Inferring biological networks by sparse identification of nonlinear dynamics. IEEE Trans. Mol. Biol. Multi-Scale Commun. 2(1), 52–63 (2016) CrossRef
11.
Mangan, N.M., Kutz, J.N., Brunton, S.L., Proctor, J.L.: Model selection for dynamical systems via sparse regression and information criteria. Proc. R. Soc. A 473(2204), 20170009 (2017)
12.
van der Meulen, F., Schauer, M., van Waaij, J.: Adaptive nonparametric drift estimation for diffusion processes using Faber-Schauder expansions. Statistical Inference for Stochastic Processes, pp. 1–26 (2017)
13.
van der Meulen, F., Schauer, M., van Zanten, H.: Reversible jump MCMC for nonparametric drift estimation for diffusion processes. Comput. Stat. Data Anal. 71, 615–632 (2014)
14.
Müller, H.G., Yao, F., et al.: Empirical dynamics for longitudinal data. Ann. Stat. 38(6), 3458–3486 (2010)
15.
Nicolau, J.: Nonparametric estimation of second-order stochastic differential equations. Econ. Theory 23(05), 880 (2007)
16.
Papaspiliopoulos, O., Roberts, G.O.: Importance sampling techniques for estimation of diffusion models. Stat. Methods Stoch. Differ. Equ. 124, 311–340 (2012)
17.
Papaspiliopoulos, O., Roberts, G.O., Stramer, O.: Data augmentation for diffusions. J. Comput. Graph. Stat. 22(3), 665–688 (2013)
18.
Quade, M., Abel, M., Kutz, J.N., Brunton, S.L.: Sparse identification of nonlinear dynamics for rapid model recovery. Chaos 28(6), 063116 (2018)
19.
Raissi, M., Karniadakis, G.E.: Hidden physics models: machine learning of nonlinear partial differential equations. J. Comput. Phys. 357, 125–141 (2018)
20.
Raissi, M., Perdikaris, P., Karniadakis, G.E.: Machine learning of linear differential equations using Gaussian processes. J. Comput. Phys. 348, 683–693 (2017)
21.
Rawat, S.: Learning governing equations for stochastic dynamical systems. Ph.D. thesis, University of California, Merced (2018). Advisor: H.S. Bhat
22.
Raziperchikolaei, R., Bhat, H.: A block coordinate descent proximal method for simultaneous filtering and parameter estimation. In: Chaudhuri, K., Salakhutdinov, R. (eds.) Proceedings of the 36th International Conference on Machine Learning. Proceedings of Machine Learning Research, Long Beach, California, USA, 09–15 June 2019, vol. 97, pp. 5380–5388. PMLR (2019)
23.
Roberts, G.O., Stramer, O.: On inference for partially observed nonlinear diffusion models using the Metropolis-Hastings algorithm. Biometrika 88(3), 603–621 (2001)
24.
Rudy, S.H., Brunton, S.L., Proctor, J.L., Kutz, J.N.: Data-driven discovery of partial differential equations. Sci. Adv. 3(4), e1602614 (2017) CrossRef
25.
Ruttor, A., Batz, P., Opper, M.: Approximate Gaussian process inference for the drift function in stochastic differential equations. In: Advances in Neural Information Processing Systems, pp. 2040–2048 (2013)
26.
Schaeffer, H., Caflisch, R., Hauck, C.D., Osher, S.: Sparse dynamics for partial differential equations. Proc. Nat. Acad. Sci. 110(17), 6634–6639 (2013)
27.
Schaeffer, H.: Learning partial differential equations via data discovery and sparse optimization. Proc. R. Soc. A: Math., Phys. Eng. Sci. 473(2197), 20160446 (2017)
28.
Schaeffer, H., Tran, G., Ward, R.: Extracting sparse high-dimensional dynamics from limited data. SIAM J. Appl. Math. 78(6), 3279–3295 (2018)
29.
Schauer, M., van der Meulen, F., van Zanten, H.: Guided proposals for simulating multi-dimensional diffusion bridges. Bernoulli 23(4A), 2917–2950 (2017)
30.
Schön, T.B., Svensson, A., Murray, L., Lindsten, F.: Probabilistic learning of nonlinear dynamical systems using sequential Monte Carlo. Mech. Syst. Signal Process. 104, 866–883 (2018) CrossRef
31.
Stuart, A.M.: Inverse problems: a Bayesian perspective. Acta Numerica 19, 451–559 (2010)
32.
Tran, G., Ward, R.: Exact recovery of chaotic systems from highly corrupted data. Multiscale Model. Simul. 15(3), 1108–1129 (2017)
33.
Verzelen, N., Tao, W., Müller, H.G.: others: Inferring stochastic dynamics from functional data. Biometrika 99(3), 533–550 (2012)
34.
Vrettas, M.D., Opper, M., Cornford, D.: Variational mean-field algorithm for efficient inference in large systems of stochastic differential equations. Phys. Rev. E 91(1), 012148 (2015)
35.
Øksendal, B.: Stochastic Differential Equations: An Introduction with Applications. Universitext, 6th edn. Springer, Heidelberg (2003). https://​doi.​org/​10.​1007/​978-3-642-14394-6