paper-with-me

홈 › Papers

Learning by solving differential equations

2025-05-19 · Benoit Dherin, Michael Munn, Hanna Mazzawi, Michael Wunder, Sourabh Medapati, Javier Gonzalvo

Modern deep learning algorithms use variations of gradient descent as their main learning methods. Gradient descent can be understood as the simplest Ordinary Differential Equation (ODE) solver; namely, the Euler method applied to the gradient flow differential equation. Since Euler, many ODE solvers have been devised that follow the gradient flow equation more precisely and more stably. Runge-Kutta (RK) methods provide a family of very powerful explicit and implicit high-order ODE solvers. However, these higher-order solvers have not found wide application in deep learning so far. In this work, we evaluate the performance of higher-order RK solvers when applied in deep learning, study their limitations, and propose ways to overcome these drawbacks. In particular, we explore how to improve their performance by naturally incorporating key ingredients of modern neural network optimizers such as preconditioning, adaptive learning rates, and momentum.

📄 PDF Abstract BibTeX arXiv:2505.13397

Code (0)

등록된 구현이 없습니다.

Tasks

Deep Learning

Similar Papers 제목 키워드 기반

FinNet: Solving Time-Independent Differential Equations with Finite Difference Neural Network

2022-02-18 · Son N. T. Tu, Thu Nguyen

Deep learning approaches for partial differential equations (PDEs) have received much attention in recent years due to their mesh-freeness and computational efficiency. However, most of the works so far have concentrated…

Computational EfficiencyDeep Learning

Hyena Neural Operator for Partial Differential Equations

2023-06-28 · Saurabh Patil, Zijie Li, Amir Barati Farimani

Numerically solving partial differential equations typically requires fine discretization to resolve necessary spatiotemporal scales, which can be computationally expensive. Recent advances in deep learning have provided…

PinnDE: Physics-Informed Neural Networks for Solving Differential Equations

2024-08-19 · Jason Matthews, Alex Bihlo

In recent years the study of deep learning for solving differential equations has grown substantially. The use of physics-informed neural networks (PINNs) and deep operator networks (DeepONets) have emerged as two of the…

One-Shot Transfer Learning of Physics-Informed Neural Networks

2021-10-21 · Shaan Desai, Marios Mattheakis, Hayden Joy, Pavlos Protopapas 외

Solving differential equations efficiently and accurately sits at the heart of progress in many areas of scientific research, from classical dynamical systems to quantum mechanics. There is a surge of interest in using P…

Transfer Learning

Legendre Deep Neural Network (LDNN) and its application for approximation of nonlinear Volterra Fredholm Hammerstein integral equations

2021-06-27 · Zeinab Hajimohammadi, Kourosh Parand, Ali Ghodsi

Various phenomena in biology, physics, and engineering are modeled by differential equations. These differential equations including partial differential equations and ordinary differential equations can be converted and…