In the current technological development, convolutional neural networks have become an important tool for computer vision tasks, especially in mobile devices. However, executing related tasks using complex neural network models often leads to high energy consumption issues. Therefore, energy modeling for neural networks becomes crucial. Through energy modeling, we can better understand the energy consumption of neural networks, and subsequently carry out targeted energy optimization to reduce the energy consumption of devices when performing tasks. This experiment selected nine feature variables from three levels of convolutional neural networks, and used five machine learning algorithms to model the energy consumption of convolutional neural networks. The five machine learning methods are Support Vector Regression (SVR), Neural Network (NN), Decision Tree (DT), Random Forest (RF), and Adaptive Boosting (AdaBoost). To select the best modeling method, this paper introduces Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) to evaluate the models. The experiment proves that Adaptive Boosting (AdaBoost) has the lowest MSE, RMSE, and MAE, therefore it is the optimal model in this experiment.

Xiaokun Qi, Tian He. Analysis and Prediction of Energy Consumption in Neural Networks Based on Machine Learning. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 4: 90-97. https://doi.org/10.25236/AJCIS.2024.070412.

[1] LIANG H, ZHANG Y, YANG P, et al. Comparison and Analysis of Prediction Models for Locomotive Traction Energy Consumption Based on the Machine Learning [J]. IEEE Access, 2023.

[2] ABDELAZIZ A, SANTOS V, DIAS M S. Convolutional Neural Network With Genetic Algorithm for Predicting Energy Consumption in Public Buildings [J]. IEEE Access, 2023, 11: 64049-69.

[3] CHEN X, CAO B, POURAMINI S. Energy cost and consumption reduction of an office building by Chaotic Satin Bowerbird Optimization Algorithm with model predictive control and artificial neural network: A case study [J]. Energy, 2023, 270: 126874.

[4] ALALI Y, HARROU F, SUN Y. Unlocking the Potential of Wastewater Treatment: Machine Learning Based Energy Consumption Prediction [J]. Water, 2023, 15(13): 2349.

[5] FARD R H, HOSSEINI S. Machine Learning algorithms for prediction of energy consumption and IoT modeling in complex networks [J]. Microprocessors and Microsystems, 2022, 89: 104423.

[6] UYAN O G, AKBAS A, GUNGOR V C. Machine learning approaches for underwater sensor network parameter prediction [J]. Ad Hoc Networks, 2023, 144: 103139.

[7] DAI X, ZHANG P, WU B, et al. Chamnet: Towards efficient network design through platform-aware model adaptation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, F, 2019 [C].

[8] BOUZIDI H, OUARNOUGHI H, NIAR S, et al. Performance prediction for convolutional neural networks on edge gpus. Proceedings of the 18th ACM International Conference on Computing Frontiers, F, 2021 [C].

[9] JUSTUS D, BRENNAN J, BONNER S, et al. Predicting the computational cost of deep learning models; proceedings of the 2018 IEEE international conference on big data (Big Data), F, 2018 [C]. IEEE.

[10] Brehler M, Camphausen L, Heidebroek B, et al. Making Machine Learning More Energy Efficient by Bringing It Closer to the Sensor [J]. IEEE Micro, 2023.

[11] Bisen D, Lilhore U K, Manoharan P, et al. A Hybrid Deep Learning Model Using CNN and K-Mean Clustering for Energy Efficient Modelling in Mobile EdgeIoT [J]. Electronics, 2023, 12(6): 1384.