Laboratory for Control, Learning, and Systems Biology

Papers by D. Ocone and E.D. Sontag

2004
  1. ▪P. Kuusela, D. Ocone, E.D. Sontag, "Learning Complexity Dimensions for a Continuous-Time Control System", SIAM J. Control Optim., vol. 43, no. 3, pp. 872–898, 2004. doipdf
    machine learning · artificial intelligence · theory of computing and complexity · VC dimension · neural networks · identifiability
    Abstract

    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.

2001
  1. ▪P. Kuusela, D. Ocone, E.D. Sontag, "Remarks on the sample complexity for linear control systems identification", In IFAC Workshop on Adaptation and Learning in Control and Signal Processing, ALCOSP2001, Cernobbio-Como, Italy, 29-31 August, 2001, pp. 431–436, 2001.
1998
  1. ▪P. Kuusela, D. Ocone, E.D. Sontag, "On the VC dimension of continuous-time linear control systems", In Proc.\ 32nd Annual Conf.\ on Information Sciences and Systems (CISS 98), Princeton, NJ, pp. 795–800, 1998.