Tensor Recovery and Machine Learning

Alex Goeßmann
Technische Universität Berlin
Institute of mathematics

There are several methods to obtain sparse models from time series of dynamical systems. While methods like SINDy are oriented on vectorial notions of sparsity, the curse of dimensionality forces us to employ tensorial analogies in cases of high-dimensional systems. In this talk, we will compare different tensor recovery methods and discuss their use in more general symbolic regression problems.


Back to Machine Learning for Physics and the Physics of Learning