Sparse Nonlinear Dynamics Models with SINDy - The Library of Candidate Nonlinearities

This course teaches learners how to choose an effective library of candidate terms for the Sparse Identification of Nonlinear Dynamics (SINDy) algorithm. The course covers extending SINDy to include control variables and bifurcation parameters, incorporating general rational functions, avoiding the curse of dimensionality using techniques like SVD, autoencoders, and tensor train formulation, and utilizing physical symmetries to constrain the library. The teaching method involves a video lecture format with chapters focusing on different aspects of building the library for SINDy. This course is intended for individuals interested in nonlinear dynamics, data science, machine learning, and control theory.