2000 character limit reached
Learning effective dynamics from data-driven stochastic systems (2205.04151v3)
Published 9 May 2022 in stat.ML and cs.LG
Abstract: Multiscale stochastic dynamical systems have been widely adopted to a variety of scientific and engineering problems due to their capability of depicting complex phenomena in many real world applications. This work is devoted to investigating the effective dynamics for slow-fast stochastic dynamical systems. Given observation data on a short-term period satisfying some unknown slow-fast stochastic systems, we propose a novel algorithm including a neural network called Auto-SDE to learn invariant slow manifold. Our approach captures the evolutionary nature of a series of time-dependent autoencoder neural networks with the loss constructed from a discretized stochastic differential equation. Our algorithm is also validated to be accurate, stable and effective through numerical experiments under various evaluation metrics.
- The lorenz-krishnamurthy slow manifold. J. Atmos. Sci., 53:1433–1437, 1996.
- D. A. Goussis. The role of slow system dynamics in predicting the degeneracy of slow invariant manifolds: The case of vdp relaxation-oscillations. Phys. D, 248:16–32, 2013.
- Empirical classification of trajectory data: An opportunity for the use of machine learning in molecular dynamics. J. Phys. Chem. B, 122 13:3230–3241, 2018.
- Chemical kinetic mechanisms and scaling of two-dimensional polymers via irreversible solution-phase reactions. J. Chem. Phys., 154 19:194901, 2021.
- Weinan E. Principles of Multiscale Modeling. Cambridge University Press, 2011.
- C. Kuehn. Multiple Time Scale Dynamics, volume 191. Springer, 2015.
- R. L. Stratonovich. Topics in the Theory of Random Noise, volume 2. CRC Press, 1967.
- L. Arnold. Random Dynamical Systems. Springer Science & Business Media, 2013.
- N. Berglund and B. Gentz. Noise-induced Phenomena in Slow-fast Dynamical Systems: A Sample-paths Approach. Springer Science & Business Media, 2006.
- G. Pavliotis and A. Stuart. Multiscale Methods: Averaging and Homogenization. Springer Science & Business Media, 2008.
- J. Duan. An Introduction to Stochastic Dynamics, volume 51. Cambridge University Press, 2015.
- D. L. Donoho. High-dimensional data analysis: The curses and blessings of dimensionality. AMS math challenges lecture, 1(2000):32, 2000.
- A survey of methods for computing (un)stable manifolds of vector fields. Int. J. Bifurcation Chaos Appl. Sci. Eng., 15:763–791, 2005.
- M. Branicki and S. Wiggins. An adaptive method for computing invariant manifolds in non-autonomous, three-dimensional dynamical systems. Phys. D, 238:1625–1657, 2009.
- S. Jain and G. Haller. How to compute invariant manifolds and their reduced dynamics in high-dimensional finite element models. Nonlinear Dyn., 107(2):1417–1450, 2022.
- Kernel methods for center manifold approximation and a weak data-based version of the center manifold theorem. Phys. D, 427:133007, 2021.
- Projecting to a slow manifold: Singularly perturbed systems and legacy codes. SIAM J. Appl. Dyn. Syst., 4:711–732, 2005.
- Invariant manifolds for stochastic partial differential equations. Ann. Probab., 31:2109–2135, 2003.
- B. Schmalfuß and K. Schneider. Invariant manifolds for random dynamical systems with slow and fast variables. J. Dyn. Differ. Equ., 20:133–164, 2008.
- Approximation of random slow manifolds and settling of inertial particles under uncertainty. J. Dyn. Differ. Equ., 27:961–979, 2012.
- Slow manifolds for multi-time-scale stochastic evolutionary systems. Communications in Mathematical Sciences, 11(1), 2012.
- RZ. Khasminskij. On the principle of averaging the itov’s stochastic differential equations. Kybernetika, 4(3):260–279, 1968.
- Stochastic averaging on graphs: Noisy duffing-van der pol equation. Stochastic and Chaotic Dynamics in the Lakes: Stochaos, 502:255–265, 2000.
- W. Wang and A. J. Roberts. Slow manifold and averaging for slow–fast stochastic differential system. J. Math. Anal. Appl., 398:822–839, 2009.
- Averaging principle for slow-fast stochastic differential equations with time dependent locally lipschitz coefficients. J. Differ. Equ., 268(6):2910–2948, 2020.
- Data-driven reduction for a class of multiscale fast-slow stochastic dynamical systems. SIAM J. Appl. Dyn. Syst., 15:1327–1351, 2016.
- Nonlinear model reduction for slow-fast stochastic systems near manifolds. ArXiv, abs/2104.02120, 2021.
- Multiscale simulations of complex systems by learning their effective dynamics. Nat. Mach. Intell., 2022.
- Predicting trajectory behaviour via machine-learned invariant manifolds. Chem. Phys. Lett., 789:139290, 2022.
- An effective manifold learning approach to parametrize data for generative modeling of biosignals. IEEE Access, 8:207112–207133, 2020.
- Data-based stochastic model reduction for the kuramoto–sivashinsky equation. Phys. D, 340:46–57, 2017.
- Deep learning to discover and predict dynamics on an inertial manifold. Phys. Rev. E, 101 6-1:062209, 2020.
- Multiscale regression on unknown manifolds. ArXiv, abs/2101.05119, 2021.
- Supervised dimensionality reduction for big data. Nat. Commun., 12, 2021.
- Manifold learning: Generalization ability and tangent proximity. International Journal of Software & Informatics, 7(3), 2013.
- N. Berglund and B. Gentz. Geometric singular perturbation theory for stochastic differential equations. J. Differ. Equ., 191:1–54, 2003.
- Z. Schuss. Theory and Applications of Stochastic Processes: An Analytical Approach, volume 170. Springer Science & Business Media, 2009.
- G. E. Hinton and R. Salakhutdinov. Reducing the dimensionality of data with neural networks. Science, 313:504 – 507, 2006.