Dresden 2020 – wissenschaftliches Programm
Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
DY: Fachverband Dynamik und Statistische Physik
DY 58: Poster: Nonlinear Dynamics; Pattern Formation; Networks; Delay Systems; Synchronization
DY 58.12: Poster
Donnerstag, 19. März 2020, 15:00–18:00, P1C
Data-driven identification of sparsely observed stochastic systems — •Dimitra Maoutsa and Manfred Opper — Technical University of Berlin, Berlin, Germany
Stochastic differential equations naturally occur in many fields in science and engineering, arising often as descriptions of systems with unresolved fast degrees of freedom. Usually, the underlying deterministic dynamics (i.e. drift function) and complete path trajectories of such systems are unknown, but instead we only have discrete time state observations at hand. Existing inference methods for such systems, either consider detailed parametric drift models, or assume densely observed trajectories for non-parametric Gaussian process drift estimation. Here, we present a data-driven approach for recovering nonlinear stochastic systems from sparse state observations. By introducing a novel method for simulating stochastic bridges consistent with the observed time series state space structure, we effectively create dense sample paths required for non-parametric drift estimation. We illustrate the power of our method on a number of simulated canonical dynamical systems, demonstrating thereby its potential and accuracy compared to existing frameworks.