Finding Strong Lottery Ticket Networks with Genetic Algorithms

Philipp Altmann, Julian Schönberger, Maximilian Zorn, Thomas Gabor

2024

Abstract

According to the Strong Lottery Ticket Hypothesis, every sufficiently large neural network with randomly initialized weights contains a sub-network which – still with its random weights – already performs as well for a given task as the trained super-network. We present the first approach based on a genetic algorithm to find such strong lottery ticket sub-networks without training or otherwise computing any gradient. We show that, for smaller instances of binary classification tasks, our evolutionary approach even produces smaller and better-performing lottery ticket networks than the state-of-the-art approach using gradient information.

Paper Citation

in Harvard Style

Altmann P., Schönberger J., Zorn M. and Gabor T. (2024). Finding Strong Lottery Ticket Networks with Genetic Algorithms. In Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: NCTA; ISBN 978-989-758-721-4, SciTePress, pages 449-460. DOI: 10.5220/0013010300003837

in Bibtex Style

@conference{ncta24,
author={Philipp Altmann and Julian Schönberger and Maximilian Zorn and Thomas Gabor},
title={Finding Strong Lottery Ticket Networks with Genetic Algorithms},
booktitle={Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: NCTA},
year={2024},
pages={449-460},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013010300003837},
isbn={978-989-758-721-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: NCTA
TI - Finding Strong Lottery Ticket Networks with Genetic Algorithms
SN - 978-989-758-721-4
AU - Altmann P.
AU - Schönberger J.
AU - Zorn M.
AU - Gabor T.
PY - 2024
SP - 449
EP - 460
DO - 10.5220/0013010300003837
PB - SciTePress