System identification through Lipschitz regularized deep neural networks
Journal of Computational Physics(2021)
摘要
•Reconstruction of a system of ODE from data using a deep neural network.•Lipschitz regularization term added in the loss function improves recovery.•The regularized model is robust to noise.•No prior knowledge on the ODE system is needed to recover the equation.•The model can handle systems of any dimension and can be used for real-world data.
更多查看译文
关键词
Machine learning,Deep learning,System identification,Ordinary differential equations,Generalization gap,Regularized network
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要