Representations of Quasi-Newton Matrices and Their Use in Limited Memory Methods

Abstract

We derive compact representations of BFGS and symmetric rank-one matrices for optimization. These representations allow us to efficiently implement limited memory methods for large constrained optimization problems. In particular, we discuss how to compute projections of limited memory matrices onto subspaces. We also present a compact representation of the matrices generated by Broyden's update for solving systems of nonlinear equations.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 06, 1992
Accession Number
ADA454688

Entities

People

  • Jorge Nocedal
  • Richard H. Byrd
  • Robert B. Schnabel

Organizations

  • University of Colorado Boulder

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Availability
  • Classification
  • Colorado
  • Computer Science
  • Computers
  • Contracts
  • Cooperation
  • Equations
  • Information Operations
  • Instructions
  • Monitoring
  • Optimization
  • Security
  • Standards

Fields of Study

  • Mathematics

Readers

  • Linear Algebra
  • Operations Research