On the Local Convergence of Pattern Search

Abstract

We examine the local convergence properties of pattern search methods, complementing the previously established global convergence properties for this class of algorithms. We show that the step-length control parameter which appears in the definition of pattern search algorithms provides a reliable asymptotic measure of first-order stationarity. This gives an analytical justification for a traditional stopping criterion for pattern search methods. Using this measure of first-order stationarity, we analyze the behavior of pattern search in the neighborhood of an isolated local minimizer. We show that a recognizable subsequence converges r-linearly to the minimizer.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 2000
Accession Number
ADA383721

Entities

People

  • Elizabeth D. Dolan
  • Robert M. Lewis
  • Virginia Torczon

Tags

Communities of Interest

  • C4I

DTIC Thesaurus Topics

  • Aeronautics
  • Algorithms
  • Computations
  • Computer Science
  • Computers
  • Contracts
  • Convergence
  • Engineering
  • Errors
  • Grids
  • Hypotheses
  • Iterations
  • Mathematics
  • Sequences
  • Standards
  • Universities
  • Virginia

Fields of Study

  • Mathematics

Readers

  • Calculus or Mathematical Analysis
  • Operations Research
  • Theoretical Analysis.