ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems
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.
Cite
Text
Wenk et al. "ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems." AAAI Conference on Artificial Intelligence, 2020. doi:10.1609/AAAI.V34I04.6106Markdown
[Wenk et al. "ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems." AAAI Conference on Artificial Intelligence, 2020.](https://mlanthology.org/aaai/2020/wenk2020aaai-odin/) doi:10.1609/AAAI.V34I04.6106BibTeX
@inproceedings{wenk2020aaai-odin,
title = {{ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems}},
author = {Wenk, Philippe and Abbati, Gabriele and Osborne, Michael A. and Schölkopf, Bernhard and Krause, Andreas and Bauer, Stefan},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2020},
pages = {6364-6371},
doi = {10.1609/AAAI.V34I04.6106},
url = {https://mlanthology.org/aaai/2020/wenk2020aaai-odin/}
}