Critical angles in polyhedral convex cones: numerical and statistical considerations

被引:12
|
作者
Gourion, Daniel [1 ]
Seeger, Alberto [1 ]
机构
[1] Univ Avignon, Dept Math, F-84000 Avignon, France
关键词
MAXIMUM-LIKELIHOOD-ESTIMATION; LOG-LINEAR MODELS; CONSTRAINTS; PAIRS;
D O I
10.1007/s10107-009-0317-2
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
This work concerns the numerical computation of critical angles in polyhedral convex cones. The set of proper critical angles is evaluated explicitly by solving a series of generalized eigenvalue problems involving the generators of the cone. The local maximal angles are identified by using a necessary condition for local maximality. The expected numbers of critical angles in random polyhedral convex cones are estimated experimentally.
引用
收藏
页码:173 / 198
页数:26
相关论文
共 50 条