TARU PUBLICATIONS
Journal of Information and Optimization Sciences cover

Journal of Information and Optimization Sciences

(Since 1980)
Hybrid ·Peer-reviewed·ISSN (Online): 2169-0103·ISSN (Print): 0252-2667

WoS  JIF 2026 : 0.4 (Q4)

Powered by:Powered by

Monthly Journal: Publishes theoretical and applied research on topics in information and optimization sciences.

Issues up to 2022 co-published with and available at:Taylor & Francis
submissions@tarupublications.com
Taru Online First Current IssueSpecial IssuesBrowse all Issues
Journal of Information and Optimization Sciences
Open Access Research Article

Optimizing energy consumption in deep learning models using pruning and quantization techniques

*

* Corresponding author · click or hover a name for details

pp. 1453–1463Vol. 45Issue 5July 2024DOI: 10.47974/JIOS-1759XML
Received:
11 Apr 2024
Published Online:
12 Aug 2024
Article type:
Research Article
Language:
EN
Article no.:
JIOS-1759
Pages:
1453–1463

Abstract

Within the past few a long time, the broad utilize of profound learning models has driven to huge enhancements in numerous regions. Be that as it may, their tall handling needs are still an issue, particularly in places with restricted assets. This think about tries to unravel these problems by proposing a better approach to utilize trimming and quantization together to form profound learning models utilize the slightest sum of energy conceivable. By carefully evacuating unnecessary links or parameters from the organize, pruning brings down the sum of work that should be done on the computer, and quantization reduces the model’s parameters to less exact representations, which assist brings down the sum of memory and work that should be done on the computer. By integrating these strategies, it conserves energy without compromising performance. In this paper, the test results demonstrate that the proposed strategy performs well with various deep learning models and datasets. Moreover this paper appears the qualities and shortcoming of the tradeoffs between show exactness and energy reserve funds, appearing how they may be utilized within the genuine world. Generally, this consider includes to the continuous work of making profound learning models that utilize less energy, which makes a difference with natural issues and makes it conceivable to utilize AI frameworks in places with restricted assets.

Keywords

Deep learningEnergy consumptionPruning techniquesQuantization techniquesComputational efficiencySustainability

Subject Classifications

Primary 00A05Secondary 97P40

References

[1] Lakshmanna, K.; Kaluni, R.; Gundluru, N.; Alzamil, Z.; Rajput, D.S.; Khan, A.A.; Haq, M.A.; Alhussen, A. A Review on Deep Learning Techniques for IoT Data. Electronics, 11, 1604 (2022).
[2] Rajput, D.S.; Reddy, T.S.K.; Raju, D.N. Investigation on Deep Learning Approach for Big Data: Applications and Challenges. Deep. Learn. Neural Netw. Concepts Methodol. Tools Appl., 11, 1604 (2020).  
[3] A. Dekhovich, D. M. J. Tax, M. H. F. Sluiter, and M. A. Bessa, “Neural network relief: a pruning algorithm based on neural activity,” arXiv, 2024. [Online]. Available: https://arxiv.org/abs/2109.10795
[4] Sharan, R.V.; Moir, T.J. An overview of applications and advancements in automatic sound recognition. Neurocomputing, 200, 22–34 (2016). 
[5] Xu, W.; Zhang, X.; Yao, L.; Xue, W.; Wei, B. A multi-view CNN-based acoustic classification system for automatic animal species identification. Ad. Hoc. Netw., 102, 102115 (2020). 
[6] Stowell, D.; Petrusková, T.; Šálek, M.; Linhart, P. Automatic acoustic identification of individuals in multiple species: Improving identification across recording conditions. J. R. Soc. Interface, 16, 20180940 (2019). 
[7] Sallam, N.M.; Saleh, A.I.; Arafat Ali, H.; Abdelsalam, M.M. An Efficient Strategy for Blood Diseases Detection Based on Grey Wolf Optimization as Feature Selection and Machine Learning Techniques. Appl. Sci., 12, 10760 (2022). 
[8] Maarif, M.R.; Listyanda, R.F.; Kang, Y.-S.; Syafrudin, M. Artificial Neural Network Training Using Structural Learning with Forgetting for Parameter Analysis of Injection Molding Quality Prediction. Information, 13, 488 (2022). 
[9] Ma, F., & Jia, R. Q. Virtual realization and geometric discriminant algorithm of the industrial robot end-effector’s position and orientation. Journal of Discrete Mathematical Sciences and Cryptography, 21(2), 471–477 (2018). 
[10] Xu, A.; Tian, M.-W.; Firouzi, B.; Alattas, K.A.; Mohammadzadeh, A.; Ghaderpour, E. A New Deep Learning Restricted Boltzmann Machine for Energy Consumption Forecasting. Sustainability, 14, 10081 (2022). 
[11] Hong, Y.; Wang, D.; Su, J.; Ren, M.; Xu, W.; Wei, Y.; Yang, Z. Short-Term Power Load Forecasting in Three Stages Based on CEEMDAN-TGA Model. Sustainability, 15, 11123 (2023). 
[12] Rashedul Islam, M., Begum, M., & Nasim Akhtar, Md. Recursive approach for multiple step-ahead software fault prediction through long short-term memory (LSTM). Journal of Discrete Mathematical Sciences and Cryptography, 25(7), 2129–2138 (2022).

Views: 130Downloads: 81Citations: 0