Selection of perturbation experiments for model discrimination

Ivayla Vatcheva; Hidde de Jong; Nicolaas Mars

Selection of perturbation experiments for model discrimination

Ivayla Vatcheva, Hidde de Jong, Nicolaas Mars

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

26 Downloads (Pure)

Abstract

It often occurs that a system can be described by several competing models. In order to distinguish among the alternative models, further information about the behavior of the system is required. One way to obtain such information is to perform suitably chosen perturbation experiments. We introduce a method for the selection of optimal perturbation experiments for discrimination among a set of dynamical models. The models are assumed to have the form of semi-quantitative differential equations. The method employs an optimization criterion based on the entropy measure of information.

Original language	English
Title of host publication	ECAI 2000
Subtitle of host publication	14th European conference on Artificial intelligence : August 20-25, 2000, Berlin, Germany
Editors	Werner Horn
Publisher	IOS
Pages	191-195
ISBN (Print)	1-5860-3013-2
Publication status	Published - 20 Aug 2000
Event	14th European Conference on Artificial Intelligence, ECAI 2000 - Berlin, Germany Duration: 20 Aug 2000 → 25 Aug 2000 Conference number: 14

Publication series

Name	Frontiers in artificial intelligence and applications
Publisher	IOS
Number	54

Conference

Conference	14th European Conference on Artificial Intelligence, ECAI 2000
Abbreviated title	ECAI
Country/Territory	Germany
City	Berlin
Period	20/08/00 → 25/08/00

Keywords

METIS-119705
IR-59859

Access to Document

Vatcheva00selection

Cite this

@inproceedings{eacf7c9dd01744978cb4642ffdd3f386,

title = "Selection of perturbation experiments for model discrimination",

abstract = "It often occurs that a system can be described by several competing models. In order to distinguish among the alternative models, further information about the behavior of the system is required. One way to obtain such information is to perform suitably chosen perturbation experiments. We introduce a method for the selection of optimal perturbation experiments for discrimination among a set of dynamical models. The models are assumed to have the form of semi-quantitative differential equations. The method employs an optimization criterion based on the entropy measure of information.",

keywords = "METIS-119705, IR-59859",

author = "Ivayla Vatcheva and {de Jong}, Hidde and Nicolaas Mars",

year = "2000",

month = aug,

day = "20",

language = "English",

isbn = "1-5860-3013-2",

series = "Frontiers in artificial intelligence and applications",

publisher = "IOS",

number = "54",

pages = "191--195",

editor = "Werner Horn",

booktitle = "ECAI 2000",

note = "14th European Conference on Artificial Intelligence, ECAI 2000, ECAI ; Conference date: 20-08-2000 Through 25-08-2000",

}

Vatcheva, I, de Jong, H & Mars, N 2000, Selection of perturbation experiments for model discrimination. in W Horn (ed.), ECAI 2000: 14th European conference on Artificial intelligence : August 20-25, 2000, Berlin, Germany. Frontiers in artificial intelligence and applications, no. 54, IOS, pp. 191-195, 14th European Conference on Artificial Intelligence, ECAI 2000, Berlin, Germany, 20/08/00.

Selection of perturbation experiments for model discrimination. / Vatcheva, Ivayla; de Jong, Hidde; Mars, Nicolaas.

ECAI 2000: 14th European conference on Artificial intelligence : August 20-25, 2000, Berlin, Germany. ed. / Werner Horn. IOS, 2000. p. 191-195 (Frontiers in artificial intelligence and applications; No. 54).

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

TY - GEN

T1 - Selection of perturbation experiments for model discrimination

AU - Vatcheva, Ivayla

AU - de Jong, Hidde

AU - Mars, Nicolaas

N1 - Conference code: 14

PY - 2000/8/20

Y1 - 2000/8/20

N2 - It often occurs that a system can be described by several competing models. In order to distinguish among the alternative models, further information about the behavior of the system is required. One way to obtain such information is to perform suitably chosen perturbation experiments. We introduce a method for the selection of optimal perturbation experiments for discrimination among a set of dynamical models. The models are assumed to have the form of semi-quantitative differential equations. The method employs an optimization criterion based on the entropy measure of information.

AB - It often occurs that a system can be described by several competing models. In order to distinguish among the alternative models, further information about the behavior of the system is required. One way to obtain such information is to perform suitably chosen perturbation experiments. We introduce a method for the selection of optimal perturbation experiments for discrimination among a set of dynamical models. The models are assumed to have the form of semi-quantitative differential equations. The method employs an optimization criterion based on the entropy measure of information.

KW - METIS-119705

KW - IR-59859

M3 - Conference contribution

SN - 1-5860-3013-2

T3 - Frontiers in artificial intelligence and applications

SP - 191

EP - 195

BT - ECAI 2000

A2 - Horn, Werner

PB - IOS

T2 - 14th European Conference on Artificial Intelligence, ECAI 2000

Y2 - 20 August 2000 through 25 August 2000

ER -

Selection of perturbation experiments for model discrimination

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this