headgraphic
loader graphic

Loading content ...

Sparse Regression and Adaptive Feature Generation for the Discovery of Dynamical Systems

Kulkarni, C.S., A. Gupta, and P.F.J. Lermusiaux, 2020. Sparse Regression and Adaptive Feature Generation for the Discovery of Dynamical Systems. In: InfoSymbiotics/DDDAS2020 Boston, 2-4 October 2020. In press.

We study the performance of sparse regression methods and propose new techniques to distill the governing equations of dynamical systems from data. We first examine the generic methodology of learning interpretable equation forms from data, proposed by Brunton et al. [2016], followed by performance of LASSO for this purpose. We then develop a new algorithm that uses the dual of LASSO optimization for higher accuracy and stability. We then obtain a novel algorithm that learns the candidate function library in a completely data-driven manner to distill the governing equations of the dynamical system. This is achieved via sequentially thresholded ridge regression (STRidge) over a orthogonal polynomial space. The performance of the methods is illustrated using the Lorenz 63 system, the quadratic Lorenz system, and a marine ecosystem model.