Two alternative schemes to update SVM approximations for the identification of explicit decision functions

Anirban Basudhar, Samy Missoum

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents a new adaptive sampling technique for the construction of explicit decision functions using support vector machine (SVM). Two approaches are used for the adaptive selection of the training samples. The first approach involves the construction of a single initial decision function using a limited number of samples, which is then up- dated by adding subsequent samples on the decision function by maximizing the minimum distance. The second approach involves the generation of competing decision functions using di®erent SVM parameters. The difference between the competing approximations provides an idea about the regions of design space with possible need for improvement. Several examples are presented to show the construction of decision functions using the proposed methods. The update schemes are validated by comparing the predicted explicit functions to actual analytical decision functions. Also, the results obtained using the two methods are compared to each other.

Original languageEnglish (US)
Title of host publication12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO
StatePublished - 2008
Event12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO - Victoria, BC, Canada
Duration: Sep 10 2008Sep 12 2008

Other

Other12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO
CountryCanada
CityVictoria, BC
Period9/10/089/12/08

Fingerprint

Support vector machines
Sampling

ASJC Scopus subject areas

  • Aerospace Engineering
  • Mechanical Engineering

Cite this

Basudhar, A., & Missoum, S. (2008). Two alternative schemes to update SVM approximations for the identification of explicit decision functions. In 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO [2008-5963]

Two alternative schemes to update SVM approximations for the identification of explicit decision functions. / Basudhar, Anirban; Missoum, Samy.

12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO. 2008. 2008-5963.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Basudhar, A & Missoum, S 2008, Two alternative schemes to update SVM approximations for the identification of explicit decision functions. in 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO., 2008-5963, 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO, Victoria, BC, Canada, 9/10/08.
Basudhar A, Missoum S. Two alternative schemes to update SVM approximations for the identification of explicit decision functions. In 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO. 2008. 2008-5963
Basudhar, Anirban ; Missoum, Samy. / Two alternative schemes to update SVM approximations for the identification of explicit decision functions. 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO. 2008.
@inproceedings{b87af89702644399b67af3a826ea0688,
title = "Two alternative schemes to update SVM approximations for the identification of explicit decision functions",
abstract = "This paper presents a new adaptive sampling technique for the construction of explicit decision functions using support vector machine (SVM). Two approaches are used for the adaptive selection of the training samples. The first approach involves the construction of a single initial decision function using a limited number of samples, which is then up- dated by adding subsequent samples on the decision function by maximizing the minimum distance. The second approach involves the generation of competing decision functions using di{\circledR}erent SVM parameters. The difference between the competing approximations provides an idea about the regions of design space with possible need for improvement. Several examples are presented to show the construction of decision functions using the proposed methods. The update schemes are validated by comparing the predicted explicit functions to actual analytical decision functions. Also, the results obtained using the two methods are compared to each other.",
author = "Anirban Basudhar and Samy Missoum",
year = "2008",
language = "English (US)",
isbn = "9781563479472",
booktitle = "12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO",

}

TY - GEN

T1 - Two alternative schemes to update SVM approximations for the identification of explicit decision functions

AU - Basudhar, Anirban

AU - Missoum, Samy

PY - 2008

Y1 - 2008

N2 - This paper presents a new adaptive sampling technique for the construction of explicit decision functions using support vector machine (SVM). Two approaches are used for the adaptive selection of the training samples. The first approach involves the construction of a single initial decision function using a limited number of samples, which is then up- dated by adding subsequent samples on the decision function by maximizing the minimum distance. The second approach involves the generation of competing decision functions using di®erent SVM parameters. The difference between the competing approximations provides an idea about the regions of design space with possible need for improvement. Several examples are presented to show the construction of decision functions using the proposed methods. The update schemes are validated by comparing the predicted explicit functions to actual analytical decision functions. Also, the results obtained using the two methods are compared to each other.

AB - This paper presents a new adaptive sampling technique for the construction of explicit decision functions using support vector machine (SVM). Two approaches are used for the adaptive selection of the training samples. The first approach involves the construction of a single initial decision function using a limited number of samples, which is then up- dated by adding subsequent samples on the decision function by maximizing the minimum distance. The second approach involves the generation of competing decision functions using di®erent SVM parameters. The difference between the competing approximations provides an idea about the regions of design space with possible need for improvement. Several examples are presented to show the construction of decision functions using the proposed methods. The update schemes are validated by comparing the predicted explicit functions to actual analytical decision functions. Also, the results obtained using the two methods are compared to each other.

UR - http://www.scopus.com/inward/record.url?scp=78049510612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049510612&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781563479472

BT - 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO

ER -