interior-point algorithms;
moment problems;
semi-infinite programming;
semidefinite programming;
nonnegative polynomials;
D O I:
暂无
中图分类号:
学科分类号:
摘要:
We generalize Nesterov's construction for the reduction of various classes of semi-infinite programming problems to the semidefinite programming form. In this way, we are able to consider ‘cones of squares’ of real-valued and matrix-valued functions as rather particular cases of a unifying abstract scheme. We also interpret from this viewpoint some results of M. Krein and A. Nudelman. This provides (in a way which probably has not been anticipated by these authors) a very powerful tool for solving various optimization problems.