Anytime Evaluation of Regression-Type Algorithms
Annamária R. Várkonyi-Kóczy, Tamás Kovácsházy, Orsolya Takács, and Csaba Benedescik
Department of Measurement and Information Systems
Budapest University of Technology and Economics, H-1521 Budapest, Magyar Tudósok körùtia 2., Hungary
Received:October 20, 2000Accepted:December 10, 2000Published:January 20, 2001
Keywords:anytime algorithms, regression-type algorithms, system modeling, complexity reduction, on-going signal processing, off-line signal processing
Regression-type algorithms are widely used for system modeling and characterization. There are applications where such characterizations are to be performed online to support control mechanisms and other decisions. In embedded autonomous systems, robustness considerations ask for techniques that, in addition to reflecting the actual state of the system and its environment, can continuously provide immediate signal processing results even in case of abrupt changes and/or temporal shortage of computational power and/or loss of some data. In other words, in such situations, actual processing should be continued to ensure appropriate performance. Consequently there is a need for robust techniques called "anytime" algorithms, which can provide short response time and be very flexible with respect to the available input information and computational power. This paper presents some considerations concerning such flexibility in the case of regression-type algorithms.
Cite this article as:A. Várkonyi-Kóczy, T. Kovácsházy, O. Takács, and C. Benedescik, “Anytime Evaluation of Regression-Type Algorithms,” J. Adv. Comput. Intell. Intell. Inform., Vol.5 No.1, pp. 2-7, 2001.Data files: