VoronoiPatches: Evaluating a New Data Augmentation Method

Steffen Illium, Gretchen Griffin, Michael Kölle, Maximilian Zorn, Jonas Nüßlein, Claudia Linnhoff-Popien

2023

Abstract

Overfitting is a problem in Convolutional Neural Networks (CNN) that causes poor generalization of models on unseen data. To remediate this problem, many new and diverse data augmentation (DA) methods have been proposed to supplement or generate more training data, and thereby increase its quality. In this work, we propose a new DA algorithm: VoronoiPatches (VP). We primarily utilize non-linear re-combination of information within an image, fragmenting and occluding small information patches. Unlike other DA methods, VP uses small convex polygon-shaped patches in a random layout to transport information around within an image. In our experiments, VP outperformed current DA methods regarding model variance and overfitting tendencies. We demonstrate DA utilizing non-linear re-combination of information within images, and non-orthogonal shapes and structures improves CNN model robustness on unseen data.

Paper Citation

in Harvard Style

Illium S., Griffin G., Kölle M., Zorn M., Nüßlein J. and Linnhoff-Popien C. (2023). VoronoiPatches: Evaluating a New Data Augmentation Method. In Proceedings of the 15th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART, ISBN 978-989-758-623-1, pages 350-357. DOI: 10.5220/0011670000003393

in Bibtex Style

@conference{icaart23,
author={Steffen Illium and Gretchen Griffin and Michael Kölle and Maximilian Zorn and Jonas Nüßlein and Claudia Linnhoff-Popien},
title={VoronoiPatches: Evaluating a New Data Augmentation Method},
booktitle={Proceedings of the 15th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART,},
year={2023},
pages={350-357},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011670000003393},
isbn={978-989-758-623-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART,
TI - VoronoiPatches: Evaluating a New Data Augmentation Method
SN - 978-989-758-623-1
AU - Illium S.
AU - Griffin G.
AU - Kölle M.
AU - Zorn M.
AU - Nüßlein J.
AU - Linnhoff-Popien C.
PY - 2023
SP - 350
EP - 357
DO - 10.5220/0011670000003393