Green AI‐Driven Concept for the Development of Cost‐Effective and Energy‐Efficient Deep Learning Method: Application in the Detection of Eimeria Parasites as a Case Study
Open Access
- 12 June 2024
- journal article
- research article
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (7)
Abstract
Although large‐scale pretrained convolutinal neural networks (CNN) models have shown impressive transfer learning capabilities, they come with drawbacks such as high energy consumption and computational cost due to their potential redundant parameters. This study presents an innovative weight‐level pruning technique that mitigates the challenges of overparameterization, and subsequently minimizes the electricity usage of such large deep learning models. The method focuses on removing redundant parameters while upholding model accuracy. This methodology is applied to classify Eimeria species parasites from fowls and rabbits. By leveraging a set of 27 pretrained CNN models with a number of parameters between 3.0M and 118.5M, the framework has identified a 4.8M‐parameter model with the highest accuracy for both animals. The model is then subjected to a systematic pruning process, resulting in an 8% reduction in parameters and a 421M reduction in floating point operations while maintaining the same classification accuracy for both fowls and rabbits. Furthermore, unlike the existing literature where two separate models are created for rabbits and fowls, this article presents a combined model with 17 classes. This approach has resulted in a CNN model with nearly 50% reduced parameter size while retaining the same accuracy of over 90%.Keywords
Funding Information
- Türkiye Bilimsel ve Teknolojik Araştırma Kurumu (Bideb‐2219‐1059B192101063)
This publication has 29 references indexed in Scilit:
- FPFS: Filter-level pruning via distance weight measuring filter similarityNeurocomputing, 2022
- Adaptive CNN filter pruning using global importance metricComputer Vision and Image Understanding, 2022
- Knowledge distillation approach towards melanoma detectionComputers in Biology and Medicine, 2022
- Lightweight convolutional neural network with knowledge distillation for cervical cells classificationBiomedical Signal Processing and Control, 2022
- CSHE: network pruning by using cluster similarity and matrix eigenvaluesInternational Journal of Machine Learning and Cybernetics, 2021
- Pruning CNN filters via quantifying the importance of deep visual representationsComputer Vision and Image Understanding, 2021
- Pruning filters with L1-norm and capped L1-norm for CNN compressionApplied Intelligence, 2020
- A dynamic CNN pruning method based on matrix similaritySignal, Image and Video Processing, 2020
- Transform domain representation-driven convolutional neural networks for skin lesion segmentationExpert Systems with Applications, 2020
- Identity Mappings in Deep Residual Networks2016