Publication Details

GPAM: Genetic Programming with Associative Memory

JŮZA Tadeáš and SEKANINA Lukáš. GPAM: Genetic Programming with Associative Memory. In: 26th European Conference on Genetic Programming (EuroGP) Held as Part of EvoStar. LNCS, vol. 13986. Cham: Springer Nature Switzerland AG, 2023, pp. 68-83. ISBN 978-3-031-29572-0. ISSN 0302-9743.
Czech title
GPAM: Genetické programování využívající asociativní paměť
Type
conference paper
Language
english
Authors
Jůza Tadeáš, Ing. (FIT BUT)
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY FIT BUT)
Keywords

Genetic programming, Associative memory, Neural network, Weight compression, Symbolic regression

Abstract

We focus on the evolutionary design of programs capable of capturing more randomness and outliers in the input data set than the standard genetic programming (GP)-based methods typically allow. We propose Genetic Programming with Associative Memory (GPAM) -- a GP-based system for symbolic regression which can utilize a small associative memory to store various data points to better approximate the original data set. The method is evaluated on five standard benchmarks in which a certain number of data points is replaced by randomly generated values. In another case study, GPAM is used as an on-chip generator capable of approximating the weights for a convolutional neural network (CNN) to reduce the access to an external weight memory. Using Cartesian genetic programming (CGP), we evolved expression-memory pairs that can generate weights of a single CNN layer.  If the associative memory contains 10% of the original weights, the weight generator evolved for a convolutional layer can approximate the original weights such that the CNN utilizing the generated weights shows less than a 1% drop in the classification accuracy on the MNIST data set. 

Published
2023
Pages
68-83
Journal
Lecture Notes in Computer Science, vol. 13986, no. 3, ISSN 0302-9743
Proceedings
26th European Conference on Genetic Programming (EuroGP) Held as Part of EvoStar
Series
LNCS
Conference
26th European Conference on Genetic Programming, Brno, CZ
ISBN
978-3-031-29572-0
Publisher
Springer Nature Switzerland AG
Place
Cham, CH
DOI
UT WoS
000999086900005
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB12860,
   author = "Tade\'{a}\v{s} J\r{u}za and Luk\'{a}\v{s} Sekanina",
   title = "GPAM: Genetic Programming with Associative Memory",
   pages = "68--83",
   booktitle = "26th European Conference on Genetic Programming (EuroGP) Held as Part of EvoStar",
   series = "LNCS",
   journal = "Lecture Notes in Computer Science",
   volume = 13986,
   number = 3,
   year = 2023,
   location = "Cham, CH",
   publisher = "Springer Nature Switzerland AG",
   ISBN = "978-3-031-29572-0",
   ISSN = "0302-9743",
   doi = "10.1007/978-3-031-29573-7\_5",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12860"
}
Back to top