Publication Details
GPAM: Genetic Programming with Associative Memory
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY)
Genetic programming, Associative memory, Neural network, Weight compression,
Symbolic regression
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.
@inproceedings{BUT185128,
author="Tadeáš {Jůza} and Lukáš {Sekanina}",
title="GPAM: Genetic Programming with Associative Memory",
booktitle="26th European Conference on Genetic Programming (EuroGP) Held as Part of EvoStar",
year="2023",
series="LNCS",
journal="Lecture Notes in Computer Science",
volume="13986",
number="3",
pages="68--83",
publisher="Springer Nature Switzerland AG",
address="Cham",
doi="10.1007/978-3-031-29573-7\{_}5",
isbn="978-3-031-29572-0",
issn="0302-9743"
}