Multiple-Cut Algorithm for Semidefinite Feasibility Problem
Xuewen Mu
Abstract
Xuewen Mu
Abstract
To solve the semidefinite feasibility problem,it is more efficient to apply the analytic center cutting plane method than to convert it a semidefinite programming problem.A multiple-cut algorithm for the problem was proposed by improving the single-cut algorithm.Its convergence was discussed.The conclusion is proved that it is a quadratic convergent method.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
To solve the semidefinite feasibility problem,it is more efficient to apply the analytic center cutting plane method than to convert it a semidefinite programming problem.A multiple-cut algorithm for the problem was proposed by improving the single-cut algorithm.Its convergence was discussed.The conclusion is proved that it is a quadratic convergent method.
Key concepts: Semidefinite embedding, Semidefinite programming, Cutting-plane method, Quadratically constrained quadratic program, Maximum cut, Convergence (economics), Algorithm, Mathematics