Number of Attention Heads vs. Number of Transformer-encoders in Computer Vision

Tomas Hrycej, Bernhard Bermeitinger, Siegfried Handschuh

2022

Abstract

Determining an appropriate number of attention heads on one hand and the number of transformer-encoders, on the other hand, is an important choice for Computer Vision (CV) tasks using the Transformer architecture. Computing experiments confirmed the expectation that the total number of parameters has to satisfy the condition of overdetermination (i.e., number of constraints significantly exceeding the number of parameters). Then, good generalization performance can be expected. This sets the boundaries within which the number of heads and the number of transformers can be chosen. If the role of context in images to be classified can be assumed to be small, it is favorable to use multiple transformers with a low number of heads (such as one or two). In classifying objects whose class may heavily depend on the context within the image (i.e., the meaning of a patch being dependent on other patches), the number of heads is equally important as that of transformers.

Paper Citation

in Harvard Style

Hrycej T., Bermeitinger B. and Handschuh S. (2022). Number of Attention Heads vs. Number of Transformer-encoders in Computer Vision. In Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2022) - Volume 1: KDIR; ISBN 978-989-758-614-9, SciTePress, pages 315-321. DOI: 10.5220/0011578000003335

in Bibtex Style

@conference{kdir22,
author={Tomas Hrycej and Bernhard Bermeitinger and Siegfried Handschuh},
title={Number of Attention Heads vs. Number of Transformer-encoders in Computer Vision},
booktitle={Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2022) - Volume 1: KDIR},
year={2022},
pages={315-321},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011578000003335},
isbn={978-989-758-614-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2022) - Volume 1: KDIR
TI - Number of Attention Heads vs. Number of Transformer-encoders in Computer Vision
SN - 978-989-758-614-9
AU - Hrycej T.
AU - Bermeitinger B.
AU - Handschuh S.
PY - 2022
SP - 315
EP - 321
DO - 10.5220/0011578000003335
PB - SciTePress