Standard

An analysis of gradient estimates in stochastic network optimization problems. / Nikolai, Krivulin.

Model-oriented data analysis : Proceedings of the 3rd international workshop in Petrodvorets, Russia, May 25-30, 1992. Physica-Verlag, 1993. p. 227-240.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Harvard

Nikolai, K 1993, An analysis of gradient estimates in stochastic network optimization problems. in Model-oriented data analysis : Proceedings of the 3rd international workshop in Petrodvorets, Russia, May 25-30, 1992. Physica-Verlag, pp. 227-240, 3rd International Workshop “Model-Oriented Data Analysis”, St. Petersburg, Russian Federation, 25/05/92. <http://pure.iiasa.ac.at/id/eprint/3705/1/XB-93-005.pdf>

APA

Nikolai, K. (1993). An analysis of gradient estimates in stochastic network optimization problems. In Model-oriented data analysis : Proceedings of the 3rd international workshop in Petrodvorets, Russia, May 25-30, 1992 (pp. 227-240). Physica-Verlag. http://pure.iiasa.ac.at/id/eprint/3705/1/XB-93-005.pdf

Vancouver

Nikolai K. An analysis of gradient estimates in stochastic network optimization problems. In Model-oriented data analysis : Proceedings of the 3rd international workshop in Petrodvorets, Russia, May 25-30, 1992. Physica-Verlag. 1993. p. 227-240

Author

Nikolai, Krivulin. / An analysis of gradient estimates in stochastic network optimization problems. Model-oriented data analysis : Proceedings of the 3rd international workshop in Petrodvorets, Russia, May 25-30, 1992. Physica-Verlag, 1993. pp. 227-240

BibTeX

@inproceedings{6a829108ff184e0d8275bbd749e15227,
title = "An analysis of gradient estimates in stochastic network optimization problems",
abstract = "Two classes of stochastic networks and their performance measures are considered. These performance measures are defined as the expected value of some random variables and cannot normally be obtained analytically as functions of network parameters. We give similar representations for the random variables to provide a useful way of analytical study of these functions and their gradients. The representations are used to obtain sufficient conditions for the gradient estimates to be unbiased. The conditions are rather general and usually met in simulation of the stochastic networks. Applications of the results are discussed and some practical algorithms of calculating unbiased estimates of the gradients are also presented.",
author = "Krivulin Nikolai",
year = "1993",
language = "English",
isbn = "3-7908-0711-7",
pages = "227--240",
booktitle = "Model-oriented data analysis",
publisher = "Physica-Verlag",
address = "Germany",
note = "3rd International Workshop “Model-Oriented Data Analysis”, MODA3 ; Conference date: 25-05-1992 Through 30-05-1992",

}

RIS

TY - GEN

T1 - An analysis of gradient estimates in stochastic network optimization problems

AU - Nikolai, Krivulin

N1 - Conference code: 3

PY - 1993

Y1 - 1993

N2 - Two classes of stochastic networks and their performance measures are considered. These performance measures are defined as the expected value of some random variables and cannot normally be obtained analytically as functions of network parameters. We give similar representations for the random variables to provide a useful way of analytical study of these functions and their gradients. The representations are used to obtain sufficient conditions for the gradient estimates to be unbiased. The conditions are rather general and usually met in simulation of the stochastic networks. Applications of the results are discussed and some practical algorithms of calculating unbiased estimates of the gradients are also presented.

AB - Two classes of stochastic networks and their performance measures are considered. These performance measures are defined as the expected value of some random variables and cannot normally be obtained analytically as functions of network parameters. We give similar representations for the random variables to provide a useful way of analytical study of these functions and their gradients. The representations are used to obtain sufficient conditions for the gradient estimates to be unbiased. The conditions are rather general and usually met in simulation of the stochastic networks. Applications of the results are discussed and some practical algorithms of calculating unbiased estimates of the gradients are also presented.

M3 - Conference contribution

SN - 3-7908-0711-7

SP - 227

EP - 240

BT - Model-oriented data analysis

PB - Physica-Verlag

T2 - 3rd International Workshop “Model-Oriented Data Analysis”

Y2 - 25 May 1992 through 30 May 1992

ER -

ID: 4406600