Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Polynomial complexity and superlinear convergence of infeasible-interior-point algorithms

Conference ·
OSTI ID:36407

Until very recently all theoretical results for interior-point-methods for linear programming have been proved under the assumption that a feasible interior point was given in advance. However, practical algorithms use starting points that lie in the interior of the region defined by the inequality constraints, but do not satisfy the equality constraints. The name infeasible-interior-point algorithm has been suggested to describe this type of method. The convergence of such algorithms was an open problem until recently, but a lot of new results have been obtained over the last couple of years. We will present a survey of the most significant contributions related to the global convergence, computational complexity, and superlinear convergence of infeasible-interior-point algorithms and will comment on their theoretical and practical consequences.

OSTI ID:
36407
Report Number(s):
CONF-9408161--
Country of Publication:
United States
Language:
English