Gene Expression Programming and Artficial Neural Network Approaches for Event Selection in High Energy Physics

Author

Teodorescu, Liliana ; Reid, Ivan D.

Author_Institution

Brunel Univ., West London

Volume

1

fYear

2006

fDate

Oct. 29 2006-Nov. 1 2006

Firstpage

593

Lastpage

598

Abstract

Gene Expression Programming is a new evolutionary algorithm found to be very efficient for solving benchmark problems from computer science. The algorithm was also successfully tested for event selection in high energy physics data analysis. This paper presents an extended event selection analysis with this algorithm, as well as a comparison of its results with those obtained with an artificial neural network. Both methods produced selection functions that allowed high classification accuracies, around 95%.

Keywords

genetic algorithms; high energy physics instrumentation computing; neural nets; Gene Expression Programming; HEP data analysis; HEP event selection; artificial neural network; evolutionary algorithm; extended event selection analysis; high energy physics; Algorithm design and analysis; Artificial neural networks; Biological cells; Data analysis; Evolutionary computation; Gene expression; Genetic programming; Neural networks; Physics; Testing; artificial neural network; classification; event selection; evolutionary algorithms; gene expression programming;

fLanguage

English

Publisher

ieee

Conference_Titel

Nuclear Science Symposium Conference Record, 2006. IEEE

Conference_Location

San Diego, CA

ISSN

1095-7863

Print_ISBN

1-4244-0560-2

Electronic_ISBN

1095-7863

Type

conf

DOI

10.1109/NSSMIC.2006.356225

Filename

4179064