arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.07104cs.LG
收藏

基于模型与物理信息的深度学习神经网络结构

Model Based and Physics Informed Deep Learning Neural Network Structures

International Science Consulting and Training (ISCT)(国际科学咨询与培训(ISCT));Zhejiang Shangfeng special blower company(浙江上风特种风机有限公司);Central South University(中南大学);Khalifa University(哈利法大学);State Key Laboratory of Airliner Integration Technology and Flight Simulation(客机集成技术与飞行模拟国家重点实验室)

Ali Mohammad-Djafari, Ning Chu, Li Wang, Caifang Cai, Liang Yu

AI总结:

针对神经网络结构选择这一开放难题,本研究结合基于模型的信号图像处理与逆问题方法,将相关网络结构设计方法分为显式解析解等五大类并举例说明,为相关网络选型与设计提供参考框架。

摘要:

神经网络(NN)已在诸多领域取得成功应用。当神经网络的结构(模型)确定后,训练阶段会通过合适的准则与优化算法(训练过程)确定模型参数,随后训练好的模型可用于预测或推理阶段(测试阶段)。由于还存在大量与优化准则、优化算法相关的超参数,在最终投入使用前需要进行验证步骤。该领域的一大难点是神经网络结构的选择:尽管已有大量“现成”网络,但针对给定的数据、信号或图像处理任务,选择或提出合适的新网络仍是一个开放性问题。本研究借助基于模型的信号与图像处理方法及逆问题方法来研究这一问题,将相关方法分为五大类:i)显式解析解,ii)变换域分解,iii)算子分解,iv)优化算法展开,v)物理信息神经网络方法(PINN,Physics Informed Neural Network),并对每一类的若干示例进行了说明。

英文摘要:

Neural Networks (NN) has been used in many areas with great success. When a NN's structure (Model) is given, during the training steps, the parameters of the model are determined using an appropriate criterion and an optimization algorithm (Training). Then, the trained model can be used for the prediction or inference step (Testing). As there are also many hyperparameters, related to the optimization criteria and optimization algorithms, a validation step is necessary before its final use. One of the great difficulties is the choice of the NN's structure. Even if there are many "on the shelf" networks, selecting or proposing a new appropriate network for a given data, signal or image processing, is still an open problem. In this work, we consider this problem using model based signal and image processing and inverse problems methods. We classify the methods in five classes, based on: i) Explicit analytical solutions, ii) Transform domain decomposition, iii) Operator Decomposition, iv) Optimization algorithms unfolding, and v) Physics Informed NN methods (PINN). Few examples in each category are explained.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑