STATISTICAL MACHINE LEARNING FOR THE COGNITIVE SELECTION OF NONLINEAR PROGRAMMING ALGORITHMS IN ENGINEERING DESIGN OPTIMIZATION.

D. A. Hoeltzel*, Wei-Hua Chieng

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

1 Scopus citations

Abstract

In order to overcome the problem of lack of generality in nonlinear programming (NLP) test problem formulation and to introduce the concept of cognitive NLP method switching, statistical machine learning has been applied to a sample data base of nonlinear programming problems. Reasonable conclusions have been drawn about an optimization problem type and a corresponding sequence of NLP solution algorithms using statistical pattern recognition applied to local (vs. global) design information. A program, referred to as OPTDEX-OLDM, with the capability of learning from statistical pattern recognition is discussed. The statistical aspects and algorithmic optimization of the nonlinear programming problem are emphasized in this discussion. A clustering process has been performed on attributes assigned to the NLP problem sample data base, and an example which describes this statistical clustering process is discussed.

Original languageEnglish
Pages67-74
Number of pages8
StatePublished - 1 Dec 1987

Fingerprint Dive into the research topics of 'STATISTICAL MACHINE LEARNING FOR THE COGNITIVE SELECTION OF NONLINEAR PROGRAMMING ALGORITHMS IN ENGINEERING DESIGN OPTIMIZATION.'. Together they form a unique fingerprint.

Cite this