ENTROPY-LIKE PROXIMAL METHODS IN CONVEX-PROGRAMMING

被引：117

作者：

IUSEM, AN ^{[1
]}

SVAITER, BF ^{[1
]}

TEBOULLE, M ^{[1
]}

机构：

[1] UNIV MARYLAND,DEPT MATH & STAT,CATONSVILLE,MD 21228

来源：

MATHEMATICS OF OPERATIONS RESEARCH | 1994年 / 19卷 / 04期

关键词：

CONVEX PROGRAMMING PROXIMAL METHODS; CONVEX STATISTICAL DISTANCES; AUGMENTED LAGRANGIANS METHODS; MODIFIED BARRIER FUNCTIONS;

D O I：

10.1287/moor.19.4.790

中图分类号：

C93 [管理学]; O22 [运筹学];

学科分类号：

070105 ; 12 ; 1201 ; 1202 ; 120202 ;

摘要：

We study an extension of the proximal method for convex programming, where the quadratic regularization kernel is substituted by a class of convex statistical distances, called phi p-divergences, which are typically entropy-like in farm. After establishing several basic properties of these quasi-distances, we present a convergence analysis of the resulting entropy-like proximal algorithm. Applying this algorithm to the dual of a convex program, we recover a wide class of nonquadratic multiplier methods and prove their convergence.

引用

页码：790 / 814

页数：25

共 27 条

[1]

ATTOUCH H, 1989, ANAL NONLINEAIRE, V6

[2] CERTAINTY EQUIVALENTS AND INFORMATION MEASURES - DUALITY AND EXTREMAL PRINCIPLES [J].