Topics in Numerical Optimization
Abstract
This grant provided support to 10 graduate students. Of these, 7 have receive their Ph.D's and another will defend her thesis before October 1, 1989. Three of these students are female US citizens. With so many graduate students involved, the research pursued has necessarily been very broad within optimization. The most exciting accomplishment is a very robust implementation of a new trust region approach to global convergence for nonlinear programming problems. The proposal for continuing this work is centered around applying this algorithm to the parallel solution of parameter identification or inverse problem for ordinary differential equations. Other work is a variant of the Kamarkar linear programming algorithm that could be of great practical significance if currently proposed extensions to nonlinear programming pan out. A unified convergence analysis for the many variants of the conjugate gradient method, a parallel direct search optimization algorithm and attendant convergence analysis, a convergence analysis for a nonsmooth trust-region method, a convergence analysis for trust-region methods for nonlinear programming, and a novel use of interactive computer graphics to obtain user performances in multi-objective optimization. Also, the PI with Professor Schnabel of Boulder completed an invited survey paper for the North Holland Handbook on Optimization. (KR)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 14, 1989
- Accession Number
- ADA213689
Entities
People
- John E. Dennis
Organizations
- Rice University