ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems

2019/02/17 by Philippe Wenk, Wenk, Philippe, Gabriele Abbati +10 · 1 citation
Computer Science · Mathematics · Physics and Astronomy · #Dynamical Systems (math.DS) #FOS: Computer and information sciences #FOS: Mathematics #Gaussian Processes and Bayesian Inference #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Model Reduction and Neural Networks #cs.LG #math.DS #stat.ML

paper · pdf · doi:10.48550/arxiv.1902.06278

Published at the Thirty-fourth AAAI Conference on Artificial Intelligence

openalex publication_date 2019/02/17 · arxiv created 2019/12/05 · arxiv updated 2019/12/06 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28

Abstract

Parameter inference in ordinary differential equations is an important problem in many applied sciences and in engineering, especially in a data-scarce setting. In this work, we introduce a novel generative modeling approach based on constrained Gaussian processes and leverage it to build a computationally and data efficient algorithm for state and parameter inference. In an extensive set of experiments, our approach outperforms the current state of the art for parameter inference both in terms of accuracy and computational cost. It also shows promising results for the much more challenging problem of model selection.

Cited by

• A cautionary tale of model misspecification and identifiability

Related

• Inference for Differential Equation Models using Relaxation via Dynamical Systems
• Learning unknown ODE models with Gaussian processes
• Model-Embedded Gaussian Process Regression for Parameter Estimation in Dynamical System
• Manifold-constrained Gaussian process inference for time-varying parameters in dynamic systems
• Fast Gaussian Process Based Gradient Matching for Parameter Identification in Systems of Nonlinear ODEs