Bare‐Bones particle Swarm optimization‐based quantization for fast and energy efficient convolutional neural networks

Tmamna, Jihene; Ayed, Emna Ben; Fourati, Rahma; Hussain, Amir; Ayed, Mounir Ben

Abstract

Neural network quantization is a critical method for reducing memory usage and computational complexity in deep learning models, making them more suitable for deployment on resource-constrained devices. In this article, we propose a method called BBPSO-Quantizer, which utilizes an enhanced Bare-Bones Particle Swarm Optimization algorithm, to address the challenging problem of mixed precision quantization of convolutional neural networks (CNNs). Our proposed algorithm leverages a new population initialization, a robust screening process, and a local search strategy to improve the search performance and guide the population towards a feasible region. Additionally, Deb's constraint handling method is incorporated to ensure that the optimized solutions satisfy the functional constraints. The effectiveness of our BBPSO-Quantizer is evaluated on various state-of-the-art CNN architectures, including VGG, DenseNet, ResNet, and MobileNetV2, using CIFAR-10, CIFAR-100, and Tiny ImageNet datasets. Comparative results demonstrate that our method delivers an excellent tradeoff between accuracy and computational efficiency.

Citation

Tmamna, J., Ayed, E. B., Fourati, R., Hussain, A., & Ayed, M. B. (2024). Bare‐Bones particle Swarm optimization‐based quantization for fast and energy efficient convolutional neural networks. Expert Systems, 41(4), Article e13522. https://doi.org/10.1111/exsy.13522

• Journal Article Type: Article
• Acceptance Date: Nov 27, 2023
• Online Publication Date: Dec 17, 2023
• Publication Date: 2024-04
• Deposit Date: Jan 12, 2024
• Publicly Available Date: Dec 18, 2024
• Journal: Expert Systems
• Print ISSN: 0266-4720
• Electronic ISSN: 1468-0394
• Publisher: Wiley
• Peer Reviewed: Peer Reviewed
• Volume: 41
• Issue: 4
• Article Number: e13522
• DOI: https://doi.org/10.1111/exsy.13522
• Keywords: Barebone PSO, energy efficient model inference, mixed precision quantization, model compression
• Public URL: http://researchrepository.napier.ac.uk/Output/3436452