Discovering a meaningful symbolic expression that explains experimental data is a fundamental challenge in many scientific fields. We present a novel, open-source computational framework called Scientist-Machine Equation Detector (SciMED), which integrates scientific discipline wisdom in a scientist-in-the-loop approach, with state-of-the-art symbolic regression (SR) methods. SciMED combines a wrapper selection method, that is based on a genetic algorithm, with automatic machine learning and two levels of SR methods. We test SciMED on five configurations of a settling sphere, with and without aerodynamic non-linear drag force, and with excessive noise in the measurements. We show that SciMED is sufficiently robust to discover the correct physically meaningful symbolic expressions from the data, and demonstrate how the integration of domain knowledge enhances its performance. Our results indicate better performance on these tasks than the state-of-the-art SR software packages , even in cases where no knowledge is integrated. Moreover, we demonstrate how SciMED can alert the user about possible missing features, unlike the majority of current SR systems.
Keren et al. - 2023 - A computational framework for physics-informed sym:/home/egpu/snap/zotero-snap/common/Zotero/storage/AIGKNXAA/Keren et al. - 2023 - A computational framework for physics-informed sym.pdf:application/pdf;Keren et al. - 2023 - A computational framework for physics-informed sym Supplementary:/home/egpu/snap/zotero-snap/common/Zotero/storage/MFHI3Z26/Keren et al. - 2023 - A computational framework for physics-informed sym.pdf:application/pdf
%0 Journal Article
%1 keren_computational_2023
%A Keren, Liron Simon
%A Liberzon, Alex
%A Lazebnik, Teddy
%D 2023
%J Scientific Reports
%K ak-symbolic-numeric domain knowledge maths physics-informed sem_ws23
%N 1
%P 1249
%R 10.1038/s41598-023-28328-2
%T A computational framework for physics-informed symbolic regression with straightforward integration of domain knowledge
%U https://www.nature.com/articles/s41598-023-28328-2
%V 13
%X Discovering a meaningful symbolic expression that explains experimental data is a fundamental challenge in many scientific fields. We present a novel, open-source computational framework called Scientist-Machine Equation Detector (SciMED), which integrates scientific discipline wisdom in a scientist-in-the-loop approach, with state-of-the-art symbolic regression (SR) methods. SciMED combines a wrapper selection method, that is based on a genetic algorithm, with automatic machine learning and two levels of SR methods. We test SciMED on five configurations of a settling sphere, with and without aerodynamic non-linear drag force, and with excessive noise in the measurements. We show that SciMED is sufficiently robust to discover the correct physically meaningful symbolic expressions from the data, and demonstrate how the integration of domain knowledge enhances its performance. Our results indicate better performance on these tasks than the state-of-the-art SR software packages , even in cases where no knowledge is integrated. Moreover, we demonstrate how SciMED can alert the user about possible missing features, unlike the majority of current SR systems.
@article{keren_computational_2023,
abstract = {Discovering a meaningful symbolic expression that explains experimental data is a fundamental challenge in many scientific fields. We present a novel, open-source computational framework called Scientist-Machine Equation Detector (SciMED), which integrates scientific discipline wisdom in a scientist-in-the-loop approach, with state-of-the-art symbolic regression (SR) methods. SciMED combines a wrapper selection method, that is based on a genetic algorithm, with automatic machine learning and two levels of SR methods. We test SciMED on five configurations of a settling sphere, with and without aerodynamic non-linear drag force, and with excessive noise in the measurements. We show that SciMED is sufficiently robust to discover the correct physically meaningful symbolic expressions from the data, and demonstrate how the integration of domain knowledge enhances its performance. Our results indicate better performance on these tasks than the state-of-the-art SR software packages , even in cases where no knowledge is integrated. Moreover, we demonstrate how SciMED can alert the user about possible missing features, unlike the majority of current SR systems.},
added-at = {2023-11-21T12:53:08.000+0100},
author = {Keren, Liron Simon and Liberzon, Alex and Lazebnik, Teddy},
biburl = {https://www.bibsonomy.org/bibtex/2cfe2db4d9eb88a0711283271e1d3bb13/martinr},
copyright = {2023 The Author(s)},
doi = {10.1038/s41598-023-28328-2},
file = {Keren et al. - 2023 - A computational framework for physics-informed sym:/home/egpu/snap/zotero-snap/common/Zotero/storage/AIGKNXAA/Keren et al. - 2023 - A computational framework for physics-informed sym.pdf:application/pdf;Keren et al. - 2023 - A computational framework for physics-informed sym Supplementary:/home/egpu/snap/zotero-snap/common/Zotero/storage/MFHI3Z26/Keren et al. - 2023 - A computational framework for physics-informed sym.pdf:application/pdf},
interhash = {b923dfdcc2f08a455b7b6ea5962dcd21},
intrahash = {cfe2db4d9eb88a0711283271e1d3bb13},
issn = {2045-2322},
journal = {Scientific Reports},
keywords = {ak-symbolic-numeric domain knowledge maths physics-informed sem_ws23},
language = {en},
month = jan,
note = {Number: 1Publisher: Nature Publishing Group},
number = 1,
pages = 1249,
timestamp = {2023-11-21T12:53:08.000+0100},
title = {A computational framework for physics-informed symbolic regression with straightforward integration of domain knowledge},
url = {https://www.nature.com/articles/s41598-023-28328-2},
urldate = {2023-11-16},
volume = 13,
year = 2023
}