Learning Complexity Dimensions for a Continuous-Time Control System

Pirkko Kuusela, Daniel Ocone, Eduardo D. Sontag
2004 SIAM Journal of Control and Optimization  
This paper takes a computational learning theory approach to a problem of linear systems identification. It is assumed that input signals have only a finite number k of frequency components, and systems to be identified have dimension no greater than n. The main result establishes that the sample complexity needed for identification scales polynomially with n and logarithmically with k.
doi:10.1137/s0363012901384302 fatcat:xeqp7u2g45bfjhgdrppaqosqji