Fast Convergence PINNs Using Pseudo-Density Embedding: A study on Solid Mechanics
Author:
Affiliation:
1. College of Computing & Data Science, Nanyang Technological University (NTU),Singapore
2. School of Aerospace Engineering, Beijing Institute of Technology,Beijing,China
3. Singapore Institute of Manufacturing Technology (SIMTech), A*STAR,Singapore
Publisher
IEEE
Link
http://xplorestaging.ieee.org/ielx8/10605128/10605229/10605316.pdf?arnumber=10605316
Reference13 articles.
1. Computer Simulation Methods
2. Deep hidden physics models: Deep learning of nonlinear partial differential equations;Raissi;The Journal of Machine Learning Research,2018
3. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
4. Lsa-pinn: Linear boundary connectivity loss for solving pdes on complex geometry;Wong
5. An expert’s guide to training physics-informed neural networks;Wang,2023
1.学者识别学者识别
2.学术分析学术分析
3.人才评估人才评估
"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370
www.globalauthorid.com
TOP
Copyright © 2019-2024 北京同舟云网络信息技术有限公司 京公网安备11010802033243号 京ICP备18003416号-3