Universal Physics-Informed Neural Networks: Symbolic Differential Operator Discovery with Sparse Data

Lena Podina, Brydon Eastman, Mohammad Kohandel
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:27948-27956, 2023.

Abstract

In this work we perform symbolic discovery of differential operators in a situation where there is sparse experimental data. This small data regime in machine learning can be made tractable by providing our algorithms with prior information about the underlying dynamics. Physics Informed Neural Networks (PINNs) have been very successful in this regime (reconstructing entire ODE solutions using only a single point or entire PDE solutions with very few measurements of the initial condition). The Universal PINN approach (UPINN) adds a neural network that learns a representation of unknown hidden terms in the differential equation. The algorithm yields both a surrogate solution to the differential equation and a black-box representation of the hidden terms. These hidden term neural networks can then be converted into symbolic equations using symbolic regression techniques like AI Feynman. In order to achieve convergence of the neural networks, we provide our algorithms with (noisy) measurements of both the initial condition as well as (synthetic) experimental data obtained at later times. We demonstrate strong performance of UPINNs even when provided with very few measurements of noisy data in both the ODE and PDE regime.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-podina23a, title = {Universal Physics-Informed Neural Networks: Symbolic Differential Operator Discovery with Sparse Data}, author = {Podina, Lena and Eastman, Brydon and Kohandel, Mohammad}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {27948--27956}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/podina23a/podina23a.pdf}, url = {https://proceedings.mlr.press/v202/podina23a.html}, abstract = {In this work we perform symbolic discovery of differential operators in a situation where there is sparse experimental data. This small data regime in machine learning can be made tractable by providing our algorithms with prior information about the underlying dynamics. Physics Informed Neural Networks (PINNs) have been very successful in this regime (reconstructing entire ODE solutions using only a single point or entire PDE solutions with very few measurements of the initial condition). The Universal PINN approach (UPINN) adds a neural network that learns a representation of unknown hidden terms in the differential equation. The algorithm yields both a surrogate solution to the differential equation and a black-box representation of the hidden terms. These hidden term neural networks can then be converted into symbolic equations using symbolic regression techniques like AI Feynman. In order to achieve convergence of the neural networks, we provide our algorithms with (noisy) measurements of both the initial condition as well as (synthetic) experimental data obtained at later times. We demonstrate strong performance of UPINNs even when provided with very few measurements of noisy data in both the ODE and PDE regime.} }
Endnote
%0 Conference Paper %T Universal Physics-Informed Neural Networks: Symbolic Differential Operator Discovery with Sparse Data %A Lena Podina %A Brydon Eastman %A Mohammad Kohandel %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-podina23a %I PMLR %P 27948--27956 %U https://proceedings.mlr.press/v202/podina23a.html %V 202 %X In this work we perform symbolic discovery of differential operators in a situation where there is sparse experimental data. This small data regime in machine learning can be made tractable by providing our algorithms with prior information about the underlying dynamics. Physics Informed Neural Networks (PINNs) have been very successful in this regime (reconstructing entire ODE solutions using only a single point or entire PDE solutions with very few measurements of the initial condition). The Universal PINN approach (UPINN) adds a neural network that learns a representation of unknown hidden terms in the differential equation. The algorithm yields both a surrogate solution to the differential equation and a black-box representation of the hidden terms. These hidden term neural networks can then be converted into symbolic equations using symbolic regression techniques like AI Feynman. In order to achieve convergence of the neural networks, we provide our algorithms with (noisy) measurements of both the initial condition as well as (synthetic) experimental data obtained at later times. We demonstrate strong performance of UPINNs even when provided with very few measurements of noisy data in both the ODE and PDE regime.
APA
Podina, L., Eastman, B. & Kohandel, M.. (2023). Universal Physics-Informed Neural Networks: Symbolic Differential Operator Discovery with Sparse Data. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:27948-27956 Available from https://proceedings.mlr.press/v202/podina23a.html.

Related Material