A strong approximation theorem for stochastic recursive algorithms

被引：14

作者：

Borkar, VS ^{[1
]}

Mitter, SK

机构：

[1] Indian Inst Sci, Dept Comp Sci & Automat, Bangalore 560012, Karnataka, India

[2] MIT, Informat & Decis Syst Lab, Cambridge, MA 02139 USA

来源：

JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS | 1999年 / 100卷 / 03期

关键词：

stochastic algorithms; approximation of stochastic differential equations; constant stepsize algorithms; asymptotic behavior;

D O I：

10.1023/A:1022630321574

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：

The constant stepsize analog of Gelfand-Mitter type discrete-time stochastic recursive algorithms is shown to track an associated stochastic differential equation in the strong sense, i.e., with respect to an appropriate divergence measure.

引用

页码：499 / 513

页数：15

共 17 条

[1] BOUNDS FOR FUNDAMENTAL SOLUTION OF A PARABOLIC EQUATION [J].

ARONSON, DG .

BULLETIN OF THE AMERICAN MATHEMATICAL SOCIETY, 1967, 73 (06) :890-&

[2]

Bhattacharya R. N., 1981, Stochastic Nonlinear Systems in Physics, Chemistry, and Biology. Proceedings of the Workshop, P86

[3]

Borkar V. S., 1986, Stochastics, V18, P17, DOI 10.1080/17442508608833398

[4] ERGODIC CONTROL OF MULTIDIMENSIONAL DIFFUSIONS .2. ADAPTIVE-CONTROL [J].

BORKAR, VS ;

GHOSH, MK .

APPLIED MATHEMATICS AND OPTIMIZATION, 1990, 21 (02) :191-220

[5]

BORKAR VS, 1996, PROBABILITY THEORY A

[6]

Cover T. M., 2005, ELEM INF THEORY, DOI 10.1002/047174882X

[7]

Deuschel J. D., 1989, LARGE DEVIATIONS

[8] METROPOLIS-TYPE ANNEALING ALGORITHMS FOR GLOBAL OPTIMIZATION IN RD [J].

GELFAND, SB ;

MITTER, SK .

SIAM JOURNAL ON CONTROL AND OPTIMIZATION, 1993, 31 (01) :111-131

[9] RECURSIVE STOCHASTIC ALGORITHMS FOR GLOBAL OPTIMIZATION IN RD [J].

GELFAND, SB ;

MITTER, SK .

SIAM JOURNAL ON CONTROL AND OPTIMIZATION, 1991, 29 (05) :999-1018

[10]

GELFAND SB, 1996, COMMUNICATION

← 1 2 →