Fitting optimal piecewise linear functions using genetic algorithms

Jennifer Pittman, C. A. Murthy

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

Constructing a model for data in R2 is a common problem in many scientific fields, including pattern recognition, computer vision, and applied mathematics. Often little is known about the process which generated the data or its statistical properties. For example, in fitting a piecewise linear model, the number of pieces, as well as the knot locations, may be unknown. Hence, the method used to build the statistical model should have few assumptions, yet, still provide a model that is optimal in some sense. Such methods can be designed through the use of genetic algorithms. In this paper, we examine the use of genetic algorithms to fit piecewise linear functions to data in R2. The number of pieces, the location of the knots, and the underlying distribution of the data are assumed to be unknown. We discuss existing methods which attempt to solve this problem and introduce a new method which employs genetic algorithms to optimize the number and location of the pieces. Experimental results are presented which demonstrate the performance of our method and compare it to the performance of several existing methods. We conclude that our method represents a valuable tool for fitting both robust and nonrobust piecewise linear functions.

Original languageEnglish (US)
Pages (from-to)701-718
Number of pages18
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume22
Issue number7
DOIs
StatePublished - Jul 1 2000

Fingerprint

Piecewise Linear Function
Genetic algorithms
Genetic Algorithm
Computer vision
Pattern recognition
Knot
Unknown
Applied mathematics
Piecewise Linear
Computer Vision
Statistical property
Statistical Model
Pattern Recognition
Linear Model
Optimise
Experimental Results
Model
Demonstrate

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

Fitting optimal piecewise linear functions using genetic algorithms. / Pittman, Jennifer; Murthy, C. A.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No. 7, 01.07.2000, p. 701-718.

Research output: Contribution to journalArticle

@article{52507519e197476383f3bd4e504cccae,
title = "Fitting optimal piecewise linear functions using genetic algorithms",
abstract = "Constructing a model for data in R2 is a common problem in many scientific fields, including pattern recognition, computer vision, and applied mathematics. Often little is known about the process which generated the data or its statistical properties. For example, in fitting a piecewise linear model, the number of pieces, as well as the knot locations, may be unknown. Hence, the method used to build the statistical model should have few assumptions, yet, still provide a model that is optimal in some sense. Such methods can be designed through the use of genetic algorithms. In this paper, we examine the use of genetic algorithms to fit piecewise linear functions to data in R2. The number of pieces, the location of the knots, and the underlying distribution of the data are assumed to be unknown. We discuss existing methods which attempt to solve this problem and introduce a new method which employs genetic algorithms to optimize the number and location of the pieces. Experimental results are presented which demonstrate the performance of our method and compare it to the performance of several existing methods. We conclude that our method represents a valuable tool for fitting both robust and nonrobust piecewise linear functions.",
author = "Jennifer Pittman and Murthy, {C. A.}",
year = "2000",
month = "7",
day = "1",
doi = "10.1109/34.865188",
language = "English (US)",
volume = "22",
pages = "701--718",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "7",

}

TY - JOUR

T1 - Fitting optimal piecewise linear functions using genetic algorithms

AU - Pittman, Jennifer

AU - Murthy, C. A.

PY - 2000/7/1

Y1 - 2000/7/1

N2 - Constructing a model for data in R2 is a common problem in many scientific fields, including pattern recognition, computer vision, and applied mathematics. Often little is known about the process which generated the data or its statistical properties. For example, in fitting a piecewise linear model, the number of pieces, as well as the knot locations, may be unknown. Hence, the method used to build the statistical model should have few assumptions, yet, still provide a model that is optimal in some sense. Such methods can be designed through the use of genetic algorithms. In this paper, we examine the use of genetic algorithms to fit piecewise linear functions to data in R2. The number of pieces, the location of the knots, and the underlying distribution of the data are assumed to be unknown. We discuss existing methods which attempt to solve this problem and introduce a new method which employs genetic algorithms to optimize the number and location of the pieces. Experimental results are presented which demonstrate the performance of our method and compare it to the performance of several existing methods. We conclude that our method represents a valuable tool for fitting both robust and nonrobust piecewise linear functions.

AB - Constructing a model for data in R2 is a common problem in many scientific fields, including pattern recognition, computer vision, and applied mathematics. Often little is known about the process which generated the data or its statistical properties. For example, in fitting a piecewise linear model, the number of pieces, as well as the knot locations, may be unknown. Hence, the method used to build the statistical model should have few assumptions, yet, still provide a model that is optimal in some sense. Such methods can be designed through the use of genetic algorithms. In this paper, we examine the use of genetic algorithms to fit piecewise linear functions to data in R2. The number of pieces, the location of the knots, and the underlying distribution of the data are assumed to be unknown. We discuss existing methods which attempt to solve this problem and introduce a new method which employs genetic algorithms to optimize the number and location of the pieces. Experimental results are presented which demonstrate the performance of our method and compare it to the performance of several existing methods. We conclude that our method represents a valuable tool for fitting both robust and nonrobust piecewise linear functions.

UR - http://www.scopus.com/inward/record.url?scp=0034230126&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034230126&partnerID=8YFLogxK

U2 - 10.1109/34.865188

DO - 10.1109/34.865188

M3 - Article

AN - SCOPUS:0034230126

VL - 22

SP - 701

EP - 718

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 7

ER -