Swap-Deep Neural Network: Incremental Inference and Learning for Embedded Systems

Taihei Asai, Koichiro Yamauchi

2024

Abstract

We propose a new architecture called “swap-deep neural network” that enables the learning and inference of large-scale artificial neural networks on edge devices with low power consumption and computational complexity. The proposed method is based on finding and integrating subnetworks from randomly initialized networks for each incremental learning phase. We demonstrate that our method achieves a performance equivalent to that of conventional deep neural networks for a variety of various classification tasks.

Paper Citation

in Harvard Style

Asai T. and Yamauchi K. (2024). Swap-Deep Neural Network: Incremental Inference and Learning for Embedded Systems. In Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-684-2, SciTePress, pages 418-427. DOI: 10.5220/0012465500003654

in Bibtex Style

@conference{icpram24,
author={Taihei Asai and Koichiro Yamauchi},
title={Swap-Deep Neural Network: Incremental Inference and Learning for Embedded Systems},
booktitle={Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2024},
pages={418-427},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012465500003654},
isbn={978-989-758-684-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - Swap-Deep Neural Network: Incremental Inference and Learning for Embedded Systems
SN - 978-989-758-684-2
AU - Asai T.
AU - Yamauchi K.
PY - 2024
SP - 418
EP - 427
DO - 10.5220/0012465500003654
PB - SciTePress