The Interior-Point Revolution in Constrained Optimization
Margaret H. Wright
Interior methods are a central, striking feature of the constrained
optimization landscape today, but it was not always so. This paper is
intended to provide a condensed, selective, unavoidably incomplete
look at classical material and recent research about interior methods.
It is based on invited survey talks at two conferences:
``High-Performance Software for Nonlinear methods. It is based on
invited survey talks at two conferences: ``High-Performance Software
for Nonlinear Optimization: Status and Perspective'', Ischia, Italy,
June 1997; and the First Pacific Rim Conference on Mathematics, Hong
Kong, January 1998.
Technical Report 98-4-09
Computing Sciences Research Center
Bell Labs, Murray Hill, New Jersey 07974