相关实验视频
Updated: Sep 9, 2025

Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
自动区分对于训练神经网络解决微分方程至关重要
Chuqi Chen1,2, Yahong Yang1, Yang Xiang1,3
1Department of Mathematics, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong.
解决部分微分方程 (PDEs) 的神经网络方法具有前景. 自动差异化 (AD) 在训练神经网络中比有限差异化 (FD) 有优势.
科学领域:
- 计算科学与工程
- 应用数学
- 用于科学计算的机器学习
背景情况:
- 神经网络越来越多地用于解决部分微分方程 (PDEs).
- 像有限差异 (FD) 这样的传统方法需要局部点来计算导数.
- 自动区分 (AD) 提供了仅使用样本点的替代方案.
研究的目的:
- 量化证明基于神经网络的PDE解答器的自动分化 (AD) 与有限差异 (FD) 方法的训练优势.
- 引入和验证一个新的度量,缩短,用于表征神经网络训练属性.
- 从培训的角度来看,比较AD和FD在解决PDEs方面的表现.
主要方法:
- 介绍培训特征的截断概念.
- 随机特征模型的实验和理论分析.
- 使用AD和FD进行双层神经网络分析.
主要成果:
- 在随机特征模型中,缩减可靠地量化剩余损失.
- 截断的作为神经网络训练速度的度量.
- 实验和理论证据表明,AD在培养神经网络的PDEs方面表现优于FD.
结论:
- 自动差异化 (AD) 与有限差异化 (FD) 方法相比,为基于神经网络的部分微分方程 (PDE) 解决者提供了一种优越的培训方法.
- 新的缩短度有效地描述了训练动态和性能.
- 这些发现支持在科学机器学习中更广泛地采用AD来解决复杂的方程.
更多相关视频
11:17Efficient Neural Differentiation using Single-Cell Culture of Human Embryonic Stem Cells
Published on: January 18, 2020
09:34A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
Published on: September 25, 2021
相关概念视频
Transmission-Line Differential Equations
Line Section Model
A circuit representing a line section of length Δx helps in understanding the transmission line parameters. The voltage V(x) and current i(x) are measured...
Linear Approximation in Frequency Domain
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
Difference Equation Solution using z-Transform
The z-transform facilitates handling delayed signals by shifting the signal in the z-domain, which corresponds to delaying the signal in the time domain, and advancing signals by similarly shifting in the...
Fundamental Mathematical Principles in Pharmacokinetics: Calculus and Graphs
On the other hand, integral calculus focuses on...
Fast Decoupled and DC Powerflow
Second Derivatives and Laplace Operator
Consider a scalar function. The curl of its...