Learning Complexity Dimensions for a Continuous-Time Control System

@article{Kuusela2004LearningCD,
  title={Learning Complexity Dimensions for a Continuous-Time Control System},
  author={Pirkko Kuusela and Daniel Ocone and Eduardo Sontag},
  journal={ArXiv},
  year={2004},
  volume={math.OC/0012163}
}
This paper takes a computational learning theory approach to a problem of linear systems identification. It is assumed that inputs are generated randomly from a known class consisting of linear combinations of k sinusoidals. The output of the system is classified at some single instant of time. The main result establishes that the number of samples needed for identification with small error and high probability, independently from the distribution of inputs, scales polynomially with n, the… 
Sample Complexity Lower Bounds for Linear System Identification
TLDR
This paper establishes problem-specific sample complexity lower bounds for linear system identification problems, and really captures the identification hardness specific to the system.
Input Classes for Identifiability of Bilinear Systems
TLDR
The main results are that step inputs are not sufficient, nor are single pulses, but the family of all pulses (of a fixed amplitude but varying widths) do suffice for identification.
Input Classes for Identification of Bilinear Systems
This paper asks what classes of input signals are sufficient in order to completely identify the input/output behavior of generic bilinear systems. The main results are that step inputs are not
Remarks on Input Classes for Identification of Bilinear Systems
TLDR
The main results are that step inputs are not sufficient, nor are single pulses, but the family of all pulses (of a fixed amplitude but varying widths) do suffice for identification.
O C ] 2 0 O ct 2 00 6 Input Classes for Identification of Bilinear Systems
This paper asks what classes of input signals are sufficient in order to completely identify the input/output behavior of generic bilinear systems. The main results are that step inputs are not

References

SHOWING 1-10 OF 35 REFERENCES
Linear systems with sign-observations
This paper deals with systems that are obtained from linear time-invariant continuous-or discrete-time devices followed by a function that just provides the sign of each output. Such systems appear
PAC-learning and asymptotic system identification theory
  • L. Ljung
  • Mathematics
    Proceedings of 35th IEEE Conference on Decision and Control
  • 1996
In this paper we discuss PAC-learning of functions from a traditional system identification perspective. The well established asymptotic theory for the identified models' properties is reviewed from
A Theory of Learning and Generalization: With Applications to Neural Networks and Control Systems
TLDR
This paper presents Vapnik-Chervonenkis and Pollard (Pseudo-) Dimensions, a model of learning based on uniform Convergence of Empirical Means, and applications to Neural Networks and Control Systems, and some Open Problems.
PAC adaptive control of linear systems
TLDR
This work proposes a learning algorithm for a special case of reinforcement learning where the environment can be described by a linear system and shows that the control law produced by the algorithm has a value close to that of an optimal policy relative to the magnitude of the initial state of the system.
Vapnik-Chervonenkis Dimension of Recurrent Neural Networks
TLDR
Lower and upper bounds for the VC dimension of recurrent networks, which are widely used in learning applications, are provided, including threshold, polynomial, piecewise-polynomial and sigmoidal functions.
Vapnik-Chervonenkis Dimension of Recurrent Neural Networks
Learning with side information: PAC learning bounds
...
1
2
3
4
...