The two-sided projected quasi-Newton method discussed in [1] is con- siderably revised by employing a trust region mutiplier strategy,introducing a nondifferentiable merit function.We develop a trust region mutiplier ...The two-sided projected quasi-Newton method discussed in [1] is con- siderably revised by employing a trust region mutiplier strategy,introducing a nondifferentiable merit function.We develop a trust region mutiplier al- gorithm for equality constained minimization.This algorithm is analyzed and global as well as local superlinear convergence theorems are proved.展开更多
Most nonliner programming problems consist of functions which are sums of unary,convex functions of linear fuctions.In this paper.we derive the duality forms of the unary oonvex optimization, and these technuqucs are ...Most nonliner programming problems consist of functions which are sums of unary,convex functions of linear fuctions.In this paper.we derive the duality forms of the unary oonvex optimization, and these technuqucs are applied to the geometric programming and minimum discrimination information problems.展开更多
文摘The two-sided projected quasi-Newton method discussed in [1] is con- siderably revised by employing a trust region mutiplier strategy,introducing a nondifferentiable merit function.We develop a trust region mutiplier al- gorithm for equality constained minimization.This algorithm is analyzed and global as well as local superlinear convergence theorems are proved.
文摘Most nonliner programming problems consist of functions which are sums of unary,convex functions of linear fuctions.In this paper.we derive the duality forms of the unary oonvex optimization, and these technuqucs are applied to the geometric programming and minimum discrimination information problems.
