Explore the Sparse Identification of Nonlinear Dynamics (SINDy) algorithm in this 27-minute video lecture. Discover how machine learning, sparse regression, and dynamical systems are combined to identify nonlinear differential equations from measurement data alone. Learn about the Lorentz Attractor, sparse regression techniques, handling noisy data, and applications to parametrized dynamics and time delay coordinates. Gain insights into the algorithm's ability to uncover governing equations from data, as presented in the 2016 PNAS paper by Brunton, Proctor, and Kutz. Access the accompanying code and delve into additional resources for a deeper understanding of this innovative approach to dynamical systems analysis.
Overview
Syllabus
Introduction
Dynamical Systems
Lorentz Attractor
Sparse Regression
Noisy Data
Example Problem
Parametrized Dynamics
Time Delay Coordinates
Taught by
Steve Brunton
Related Courses
-
Sparse Identification of Nonlinear Dynamics - Sparse Machine Learning Models 5 Years Later
-
Identifying Nonlinear Dynamics with High Confidence from Sparse Data
-
Dynamical Systems
-
Data-Driven Dynamical Systems with Machine Learning
-
Sparse Nonlinear Models for Fluid Dynamics with Machine Learning and Optimization
-
Learn One Size to Infer All: Exploiting Symmetries in Dynamical Systems Using Scalable Neural Networks
