Logarithmic Indicators and the Identification of Subgroups of Variables in Interior-Point Methods
Abstract
The identification of certain groups of variables in optimization problems is an important issue and can be used to computational advantage. In this paper new logarithmic indicators are introduced. It is demonstrated that the logarithmic Tapia indicators have superior ability for identifying several subgroups of variables in the context of primal-dual interior-point methods.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1993
- Accession Number
- ADA444980
Entities
People
- A. S. El-bakry
- Richard A. Tapia
- Yinglong Zhang
Organizations
- Rice University