mlsanityData to 14 Oct 2026

Recognition Models to Learn Dynamics from Partial Observations with Neural ODEs

arXiv 2205.12550 ↗First posted 25 May 2022eess.SY cs.LG cs.SYcited by 8

Not on the map. Why?

Abstract (arXiv)

Identifying dynamical systems from experimental data is a notably difficult task. Prior knowledge generally helps, but the extent of this knowledge varies with the application, and customized models are often needed. Neural ordinary differential equations can be written as a flexible framework for system identification and can incorporate a broad spectrum of physical insight, giving physical interpretability to the resulting latent space. In the case of partial observations, however, the data points cannot directly be mapped to the latent state of the ODE. Hence, we propose to design recognition models, in particular inspired by nonlinear observer theory, to link the partial observations to the latent state. We demonstrate the performance of the proposed approach on numerical simulations and on an experimental dataset from a robotic exoskeleton.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 35 of its 73 references that are in the map. First posted on arXiv 25 May 2022.

Builds on

35 of 73 references.

References · 35

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2018Neural Ordinary Differential Equationscited by 3,779
  3. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,465
  4. 2019Hamiltonian Neural Networkscited by 712
  5. 2021Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layerscited by 608
  6. 2020Neural Controlled Differential Equations for Irregular Time Seriescited by 506
  7. 2017Deep learning for universal linear embeddings of nonlinear dynamicscited by 494
  8. 2020Lagrangian Neural Networkscited by 398
  9. 2019Augmented Neural ODEscited by 397
  10. 2019GRU-ODE-Bayes: Continuous modeling of sporadically-observed time seriescited by 231
  11. 2019Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learningcited by 212
  12. 2019Symplectic ODE-Net: Learning Hamiltonian Dynamics with Controlcited by 184
23 more
  1. 2019Latent ODEs for Irregularly-Sampled Time Seriescited by 183
  2. 2020Dissecting Neural ODEscited by 153
  3. 2020Learning Stable Deep Dynamics Modelscited by 111
  4. 2020Augmenting Physical Models with Deep Networks for Complex Dynamics Forecastingcited by 95
  5. 2020On Second Order Behaviour in Augmented Neural ODEscited by 68
  6. 2018Probabilistic Recurrent State-Space Modelscited by 64
  7. 2022Discovering Governing Equations from Partial Measurements with Deep Delay Autoencoderscited by 58
  8. 2021Neural ODE Processescited by 57
  9. 2020Benchmarking Energy-Conserving Neural Networks for Learning Dynamics from Datacited by 40
  10. 2017Identification of Gaussian Process State Space Modelscited by 31
  11. 2021Constructing Neural Network-Based Models for Simulating Dynamical Systemscited by 31
  12. 2020Stable Neural Flowscited by 25
  13. 2021Learning Stable Deep Dynamics Models for Partially Observed or Delayed Dynamical Systemscited by 17
  14. 2020Bayesian System ID: Optimal management of parameter, model, and measurement uncertaintycited by 16
  15. 2021Beyond Predictions in Neural ODEs: Identification and Interventionscited by 15
  16. 2022Towards gain tuning for numerical KKL observerscited by 14
  17. 2021Deep KKL: Data-driven Output Prediction for Non-Linear Systemscited by 10
  18. 2021Structure-preserving Gaussian Process Dynamicscited by 9
  19. 2021NeuralPDE: Modelling Dynamical Systems from Datacited by 9
  20. 2021Weak Form Generalized Hamiltonian Learningcited by 8
  21. 2021Physics-enhanced Neural Networks in the Small Data Regimecited by 3
  22. 2021Modular Neural Ordinary Differential Equationscited by 2
  23. 2021Characteristic Neural Ordinary Differential Equationscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 2; 2024: 1; 2025: 1; 2026: 3.

8 papers.

  1. 2024Learning Physics Informed Neural ODEs With Partial Measurementscited by 9
  2. 2022Learning Robust State Observers using Neural ODEs (longer version)cited by 7
  3. 2023Some of the variables, some of the parameters, some of the times, with some physics known: Identification with partial informationcited by 4
  4. 2025Data-efficient Kernel Methods for Learning Hamiltonian Systemscited by 2
  5. 2023Learning Hybrid Dynamics Models With Simulator-Informed Latent Statescited by 1
  6. 2026History-informed Lagrangian Neural Networkscited by 1
  7. 2026Learning partially observed systems with neural Hamiltonian ordinary differential equationscited by 0
  8. 2026RTS Smoother-Guided Learning of Physics-Based Neural Differential Modelscited by 0