Convergence Issues in Stochastic Optimization

Pierre Carpentier; Jean Philippe Chancelier; Guy Cohen; Michel De Lara

doi:10.1007/978-3-319-18138-7_8

Convergence Issues in Stochastic Optimization

Pierre Carpentier
, Jean Philippe Chancelier
, Guy Cohen
, Michel De Lara

École des ponts

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

Abstract

In this chapter, we prove a convergence theorem for appropriate notions of noise and information convergence in closed-loop stochastic optimization problems. The main Theorem is based on epi-convergence results, which take into account constraints defined by characteristic functions of sets.

Original language	English
Title of host publication	Probability Theory and Stochastic Modelling
Publisher	Springer Nature
Pages	211-252
Number of pages	42
DOIs	https://doi.org/10.1007/978-3-319-18138-7_8
Publication status	Published - 1 Jan 2015

Publication series

Name	Probability Theory and Stochastic Modelling
Volume	75
ISSN (Print)	2199-3130
ISSN (Electronic)	2199-3149

Keywords

Convergence Notions
Measurement Constraints
Mosco Convergence
Normal Integrand
Random Variable Approximations

Access to Document

10.1007/978-3-319-18138-7_8

Cite this

@inbook{4150db51749244c18a1142959f4d03a8,

title = "Convergence Issues in Stochastic Optimization",

abstract = "In this chapter, we prove a convergence theorem for appropriate notions of noise and information convergence in closed-loop stochastic optimization problems. The main Theorem is based on epi-convergence results, which take into account constraints defined by characteristic functions of sets.",

keywords = "Convergence Notions, Measurement Constraints, Mosco Convergence, Normal Integrand, Random Variable Approximations",

author = "Pierre Carpentier and Chancelier, \{Jean Philippe\} and Guy Cohen and \{De Lara\}, Michel",

note = "Publisher Copyright: {\textcopyright} 2015, Springer International Publishing Switzerland.",

year = "2015",

month = jan,

day = "1",

doi = "10.1007/978-3-319-18138-7\_8",

language = "English",

series = "Probability Theory and Stochastic Modelling",

publisher = "Springer Nature",

pages = "211--252",

booktitle = "Probability Theory and Stochastic Modelling",

}

TY - CHAP

T1 - Convergence Issues in Stochastic Optimization

AU - Carpentier, Pierre

AU - Chancelier, Jean Philippe

AU - Cohen, Guy

AU - De Lara, Michel

PY - 2015/1/1

Y1 - 2015/1/1

N2 - In this chapter, we prove a convergence theorem for appropriate notions of noise and information convergence in closed-loop stochastic optimization problems. The main Theorem is based on epi-convergence results, which take into account constraints defined by characteristic functions of sets.

AB - In this chapter, we prove a convergence theorem for appropriate notions of noise and information convergence in closed-loop stochastic optimization problems. The main Theorem is based on epi-convergence results, which take into account constraints defined by characteristic functions of sets.

KW - Convergence Notions

KW - Measurement Constraints

KW - Mosco Convergence

KW - Normal Integrand

KW - Random Variable Approximations

UR - https://www.scopus.com/pages/publications/85130817851

U2 - 10.1007/978-3-319-18138-7_8

DO - 10.1007/978-3-319-18138-7_8

M3 - Chapter

AN - SCOPUS:85130817851

T3 - Probability Theory and Stochastic Modelling

SP - 211

EP - 252

BT - Probability Theory and Stochastic Modelling

PB - Springer Nature

ER -

Convergence Issues in Stochastic Optimization

Abstract

Publication series

Keywords

Access to Document

Other files and links

Fingerprint

Cite this