.png){kind=icon}
返回
Discovering sparse interpretable dynamics from partial observations
.png){kind=icon}
OA
AI
DOI：10.1038/s42005-022-00987-z.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Identifying the governing equations of a nonlinear dynamical system is key to both understanding the physical features of the system and constructing an accurate model of the dynamics that generalizes well beyond the available data. Achieving this kind of interpretable system identification is even more difficult for partially observed systems. We propose a machine learning framework for discovering the governing equations of a dynamical system using only partial observations, combining an encoder for state reconstruction with a sparse symbolic model. The entire architecture is trained end-to-end by matching the higher-order symbolic time derivatives of the sparse symbolic model with finite difference estimates from the data. Our tests show that this method can successfully reconstruct the full system state and identify the equations of motion governing the underlying dynamics for a variety of ordinary differential equation (ODE) and partial differential equation (PDE) systems. Nonlinear dynamical systems are ubiquitous in nature and play an essential role in science, from providing models for the weather forecast to describing the chaotic behavior of plasma in nuclear reactors. This paper introduces an artificial intelligence framework that can learn the correct equations of motion for nonlinear systems from incomplete data, and opens up the door to applying interpretable machine learning techniques on a wide range of applications in the field of nonlinear dynamics.
Keyword:
NEURAL-NETWORKS
IDENTIFICATION
PREDATOR
期刊
IF:
5.8
论文数:
2.8K
被引数:
9.2K
机构
暂无机构信息
引用论文
分享收藏Physics-incorporated convolutional recurrent neural networks for source identification and forecasting of dynamical systems结合物理的卷积递归神经网络用于动力系统的源识别和预测.png)
NEURAL NETWORKS
OAAI
NEURAL NETWORKSIF6.3
OAAI分享收藏Can rationing increase welfare? Theory and an application to India's ration shop system配给制能增加福利吗？理论及其在印度配给店系统中的应用OAAI
IF0
OAAI分享收藏Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations物理信息神经网络: 一种用于解决涉及非线性偏微分方程的正反问题的深度学习框架OAAI
OAAI分享收藏Physics-informed learning of governing equations from scarce data从稀缺数据中学习控制方程的物理知识NATURE COMMUNICATIONSOAAI
NATURE COMMUNICATIONSIF15.7
OAAI分享收藏Discovering governing equations from data by sparse identification of nonlinear dynamical systems通过非线性动力系统的稀疏识别从数据中发现控制方程OAAI
OAAI分享收藏
查看更多内容