Lightweight CNN Design Based on Mixup Data Augmentation and Network Pruning

Yuxuan Li

doi:10.54254/2755-2721/2025.AST26567

Applied and Computational EngineeringOpen access

Lightweight CNN Design Based on Mixup Data Augmentation and Network Pruning

Research Article

Open Access

Lightweight CNN Design Based on Mixup Data Augmentation and Network Pruning

Yuxuan Li ^1*

¹ Institute of East China University of Science and Technology, Shanghai, China; Department of Computer Science, ECUST, Shanghai, China

^*Corresponding author: 3027435227@qq.com

Published on 9 September 2025

ACE Vol.185

ISSN (Print): 2755-273X

ISSN (Online): 2755-2721

ISBN (Print): 978-1-80590-369-7

ISBN (Online): 978-1-80590-370-3

Download Cover

Abstract

Convolutional Neural Networks (CNNs) have achieved remarkable success in image classification and other vision tasks in recent years. However, their large model size and computational complexity hinder their application in mobile terminals and embedded devices. To address this issue, this paper proposes a lightweight CNN design method that combines Mixup data augmentation and network pruning. The method aims to balance the trade-off between model compression and performance preservation, achieving the maximum model compression while maintaining as much of the original performance as possible. Using the FashionMNIST dataset as the experimental platform, a classification model based on a simplified LeNet structure is constructed. The model is evaluated under four different settings: the standard model, the Mixup-augmented model, the pruned sparse model, and the collaborative model integrating both Mixup augmentation and pruning. The experimental results show that Mixup enhances the model's generalization ability and robustness, pruning significantly reduces the number of parameters, and the combination of both achieves superior lightweight performance while preserving accuracy. This study demonstrates the effectiveness of Mixup and pruning techniques in collaborative optimization and proposes practical optimization strategies for deploying lightweight neural networks in resource-constrained environments.

Keywords:

Lightweight Convolutional Neural Network, Mixup Data Augmentation, Network Pruning, Model Compression, Image Classification, FashionMNIST

View PDF

References

[1]. Kang M, Kim S. Guidedmixup: an efficient mixup strategy guided by saliency maps [C]//Proceedings of the AAAI conference on artificial intelligence. 2023, 37(1): 1096-1104.

[2]. Molchanov P, Mallya A, Tyree S, et al. Importance estimation for neural network pruning [C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2019: 11264-11272.

[3]. Xiao H, Rasul K, Vollgraf R. Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms [J]. arXiv preprint arXiv: 1708.07747, 2017.

[4]. Bouti A, Mahraz M A, Riffi J, et al. A robust system for road sign detection and classification using LeNet architecture based on convolutional neural network [J]. Soft Computing, 2020, 24(9): 6721-6733.

[5]. Maharana K, Mondal S, Nemade B. A review: Data pre-processing and data augmentation techniques [J]. Global Transitions Proceedings, 2022, 3(1): 91-99.

[6]. Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks [J]. Communications of the ACM, 2020, 63(11): 139-144.

[7]. Kusner M J, Paige B, Hernández-Lobato J M. Grammar variational autoencoder [C]//International conference on machine learning. PMLR, 2017: 1945-1954.

[8]. Chapelle O, Weston J, Bottou L, et al. Vicinal risk minimization [J]. Advances in neural information processing systems, 2000, 13.

[9]. Zagoruyko S, Komodakis N. Wide residual networks [J]. arXiv preprint arXiv: 1605.07146, 2016.

[10]. Zhong L, Wan F, Chen R, et al. Blockpruner: Fine-grained pruning for large language models [J]. arXiv preprint arXiv: 2406.10594, 2024.

[11]. Anwar S, Hwang K, Sung W. Structured pruning of deep convolutional neural networks [J]. ACM Journal on Emerging Technologies in Computing Systems (JETC), 2017, 13(3): 1-18.

[12]. Jordao A, Lie M, Schwartz W R. Discriminative layer pruning for convolutional neural networks [J]. IEEE Journal of Selected Topics in Signal Processing, 2020, 14(4): 828-837.

[13]. Abdelkhalik H, Arafa Y, Santhi N, et al. Demystifying the nvidia ampere architecture through microbenchmarking and instruction-level analysis [C]//2022 IEEE High Performance Extreme Computing Conference (HPEC). Ieee, 2022: 1-8.

[14]. Shah A, Shao M. Deep Compression with Adversarial Robustness Via Decision Boundary Smoothing [J]. 2025.

[15]. Gale T, Elsen E, Hooker S. The state of sparsity in deep neural networks [J]. arXiv preprint arXiv: 1902.09574, 2019