Conditions of optimal classification for piecewise affine regression

Authors:
Giancarlo Ferrari-Trecate;Michael Schinkel
Affiliations:
INRIA, Le Chesnay Cedex, France;Department of Mechanical Engineering, Center for Systems and Control, University of Glasgow, Glasgow, Scotland
Venue:
HSCC'03 Proceedings of the 6th international conference on Hybrid systems: computation and control
Year:
2003

Citing 4
Cited 3

Identification of non-linear system structure and parameters using regime decomposition

Automatica (Journal of IFAC)
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Nonlinear filters based on combinations of piecewise polynomialswith compact support

IEEE Transactions on Signal Processing
A clustering technique for the identification of piecewise affine systems

Automatica (Journal of IFAC)

Identification of switched linear systems via sparse optimization

Automatica (Journal of IFAC)
Comparison of four procedures for the identification of hybrid systems

HSCC'05 Proceedings of the 8th international conference on Hybrid Systems: computation and control
Identification of piecewise affine systems based on statistical clustering technique

Automatica (Journal of IFAC)

Quantified Score

Hi-index	0.00

Visualization

Abstract

We consider regression problems with piecewise affine maps. In particular, we focus on the sub-problem of classifying the datapoints, i.e. correctly attributing a datapoint to the affine submodel that most likely generated it. Then, we analyze the regression algorithm proposed in [4,3] and show that, under suitable assumptions on the dataset and the weights of the classification procedure, optimal classification can be guaranteed in presence of bounded noise. We also relax such assumptions by introducing and characterizing the property of weakly optimal classification. Finally, by elaborating on these concepts, we propose a procedure for detecting, a posteriori, misclassified datapoints.