“Take a look at [2] for a more detailed description.”
Universal Differential Equations for Scientific Machine Learning
提出通用微分方程框架,把物理模型与机器学习统一进 SciML 生态
介绍 SciML 软件生态,用来把物理定律和科学模型的信息与数据驱动的机器学习混合起来,并提出通用微分方程(UDE)这一统一框架。从自动发现生物机制到求解高维 Hamilton-Jacobi-Bellman 方程等多种应用,都能用 UDE 表述并高效处理;工具可处理随机性、延迟与隐式约束,针对刚性方程做过稳定化,并兼容分布式并行和 GPU。
评论者(包括作者本人)用它指代把已知科学结构与机器学习混合的通用框架,涉及方程自动发现、用傅里叶级数替代神经网络有时更快,以及神经网络与符号回归结合。
HN 书评
推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及“[1] Universal Differential Equations for Scientific Machine Learning, Universal Differential Equations for Scientific Machine Learning”
“A lot of these methods here utilize the universal differential equation framework described here: Universal Differential Equations for Scientific Machine Learning”
“Along the lines of this sort of stuff [2].”
“Yes, and our recent work Universal Differential Equations for Scientific Machine Learning gives a fairly general form for how to mix known scientific structural knowledge directly with machine learning.”
“In the universal differential equations paper, we demonstrate that some examples of automatic equation discovery are faster using Fourier series than NNs”
看全部 7 条 →
“demonstrates these cases where neural networks in differential equations are mixed with symbolic regression”
- Universal Differential Equations for Scientific Machine Learning ↗ChrisRackauckas · 2020-01-14 · 6 分
- Universal Differential Equations for ScientificMachine Learning ↗tu7001 · 2020-05-04 · 1 分
- Universal Differential Equations for Scientific Machine Learning ↗formalsystem · 2020-05-17 · 3 分
- Universal Differential Equations for Scientific Machine Learning ↗sriku · 2020-08-06 · 2 分