Learning of Linear Dynamical Systems as a Non-Commutative Polynomial Optimization Problem

Abstract

There has been much recent progress in forecasting the next observation of a linear dynamical system (LDS), which is known as the improper learning, as well as in the estimation of its system matrices, which is known as the proper learning of LDS. We present an approach to proper learning of LDS, which in spite of the non-convexity of the problem, guarantees global convergence of numerical solutions to a least-squares estimator. We present promising computational results.

14 pages, 4 figures; retitled to reflect the title of the the published version

References in corpus (5)

• Chordal-TSSOS: a moment-SOS hierarchy that exploits term sparsity with chordal extension
• Sparse Noncommutative Polynomial Optimization
• Fairness in Forecasting of Observations of Linear Dynamical Systems
• Quantum open system identification via global optimization: Optimally accurate Markovian models of open systems from time-series data
• Learning of Linear Dynamical Systems as a Non-Commutative Polynomial Optimization Problem

Cited by in corpus (1)

• Learning of Linear Dynamical Systems as a Non-Commutative Polynomial Optimization Problem