Partial differential equations meet deep neural networks: A survey

S Huang, W Feng, C Tang, J Lv - arXiv preprint arXiv:2211.05567, 2022 - arxiv.org
Many problems in science and engineering can be represented by a set of partial differential
equations (PDEs) through mathematical modeling. Mechanism-based computation following …

Multi-Scale Simulation of Complex Systems: A Perspective of Integrating Knowledge and Data

H Wang, H Yan, C Rong, Y Yuan, F Jiang… - ACM Computing …, 2024 - dl.acm.org
Complex system simulation has been playing an irreplaceable role in understanding,
predicting, and controlling diverse complex systems. In the past few decades, the multi-scale …

Deep learning of parameterized equations with applications to uncertainty quantification

T Qin, Z Chen, JD Jakeman… - International Journal for …, 2021 - dl.begellhouse.com
We propose a learning algorithm for discovering unknown parameterized dynamical
systems by using observational data of the state variables. Our method is built upon and …

Microfounded tax revenue forecast model with heterogeneous population and genetic algorithm approach

A Alexi, T Lazebnik, L Shami - Computational Economics, 2024 - Springer
The ability of governments to accurately forecast tax revenues is essential for the successful
implementation of fiscal programs. However, forecasting state government tax revenues …

Fast fitting of neural ordinary differential equations by Bayesian neural gradient matching to infer ecological interactions from time‐series data

W Bonnaffé, T Coulson - Methods in Ecology and Evolution, 2023 - Wiley Online Library
Inferring ecological interactions is hard because we often lack suitable parametric
representations to portray them. Neural ordinary differential equations (NODEs) provide a …

MINN: Learning the dynamics of differential-algebraic equations and application to battery modeling

Y Huang, C Zou, Y Li, T Wik - IEEE Transactions on Pattern …, 2024 - ieeexplore.ieee.org
The concept of integrating physics-based and data-driven approaches has become popular
for modeling sustainable energy systems. However, the existing literature mainly focuses on …

Entropy structure informed learning for solving inverse problems of differential equations

Y Jiang, W Yang, Y Zhu, L Hong - Chaos, Solitons & Fractals, 2023 - Elsevier
Entropy, since its first discovery by Ludwig Boltzmann in 1877, has been widely applied in
diverse disciplines, including thermodynamics, continuum mechanics, mathematical …

Implementation and (Inverse Modified) Error Analysis for Implicitly Templated ODE-Nets

A Zhu, T Bertalan, B Zhu, Y Tang, IG Kevrekidis - SIAM Journal on Applied …, 2024 - SIAM
We focus on learning unknown dynamics from data using ODE-nets templated on implicit
numerical initial value problem solvers. First, we perform inverse modified error analysis of …

PolyODENet: Deriving mass-action rate equations from incomplete transient kinetics data

Q Wu, T Avanesian, X Qu, H Van Dam - The Journal of Chemical …, 2022 - pubs.aip.org
Kinetics of a reaction network that follows mass-action rate laws can be described with a
system of ordinary differential equations (ODEs) with polynomial right-hand side. However, it …

Weak collocation regression method: Fast reveal hidden stochastic dynamics from high-dimensional aggregate data

L Lu, Z Zeng, Y Jiang, Y Zhu, P Hu - Journal of Computational Physics, 2024 - Elsevier
Revealing hidden dynamics from the stochastic data is a challenging problem as the
randomness takes part in the evolution of the data. The problem becomes exceedingly hard …