Interior-Point Methods for Convex Programming
Abstract
This work is concerned with generalized convex programming problems, where the objective and also the constraints belong to a certain class of convex functions. It examines the relationship of two conditions for generalized convex programming--self concordance and a relative Lipschitz condition--and gives an outline for a short and simple analysis of an interior-point method for generalized convex programming. It generalizes ellipsoidal approximations for the feasible set, and in the special case of a nondegenerate linear program it establishes a uniform bound on the condition number of the matrices occurring when the iterates remain near the path of centers.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1990
- Accession Number
- ADA231372
Entities
People
- Florian Jarre
Organizations
- Stanford University