Learning Dynamical Systems from Partial Observations

Abstract

We consider the problem of forecasting complex, nonlinear space-time processes when observations provide only partial information of on the system's state. We propose a natural data-driven framework, where the system's dynamics are modelled by an unknown time-varying differential equation, and the evolution term is estimated from the data, using a neural network. Any future state can then be computed by placing the associated differential equation in an ODE solver. We first evaluate our approach on shallow water and Euler simulations. We find that our method not only demonstrates high quality long-term forecasts, but also learns to produce hidden states closely resembling the true states of the system, without direct supervision on the latter. Additional experiments conducted on challenging, state of the art ocean simulations further validate our findings, while exhibiting notable improvements over classical baselines.

References in corpus (3)

• Deep Learning for Physical Processes: Incorporating Prior Scientific Knowledge
• Bilinear residual Neural Network for the identification and forecasting of dynamical systems
• Optimal Unsupervised Domain Translation

Cited by in corpus (21)

• Symplectic ODE-Net: Learning Hamiltonian Dynamics with Control
• Hybrid FEM-NN models: Combining artificial neural networks with the finite element method
• Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting
• A review of radar-based nowcasting of precipitation and applicable machine learning techniques
• Parameterized Neural Ordinary Differential Equations: Applications to Computational Physics Problems
• Dissipative SymODEN: Encoding Hamiltonian Dynamics with Dissipation and Control into Deep Learning
• Deep reconstruction of strange attractors from time series
• Adaptive Checkpoint Adjoint Method for Gradient Estimation in Neural ODE
• Bridging Physics-based and Data-driven modeling for Learning Dynamical Systems
• Learning Generalized Quasi-Geostrophic Models Using Deep Neural Numerical Models
• Identifying nonlinear dynamical systems with multiple time scales and long-range dependencies
• Neural Dynamical Systems: Balancing Structure and Flexibility in Physical Prediction
• Bridging the Gap: Machine Learning to Resolve Improperly Modeled Dynamics
• Variational Deep Learning for the Identification and Reconstruction of Chaotic and Stochastic Dynamical Systems from Noisy and Partial Observations
• Meta-Learning Dynamics Forecasting Using Task Inference
• DynaBench: A benchmark dataset for learning dynamical systems from low-resolution data
• Deep learning reveals hidden interactions in complex systems
• Reconstructing a dynamical system and forecasting time series by self-consistent deep learning
• Stability of implicit neural networks for long-term forecasting in dynamical systems
• Probabilistic sequential matrix factorization
• Physics-based polynomial neural networks for one-shot learning of dynamical systems from one or a few samples