Optimization of Layer Skipping and Frequency Scaling for Convolutional Neural Networks under Latency Constraint
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910899758956544 |
|---|---|
| author | Chan, Minh David Thao Zhao, Ruoyu Jia, Yukuan Mao, Ruiqing Zhou, Sheng |
| author_facet | Chan, Minh David Thao Zhao, Ruoyu Jia, Yukuan Mao, Ruiqing Zhou, Sheng |
| contents | The energy consumption of Convolutional Neural Networks (CNNs) is a critical factor in deploying deep learning models on resource-limited equipment such as mobile devices and autonomous vehicles. We propose an approach involving Proportional Layer Skipping (PLS) and Frequency Scaling (FS). Layer skipping reduces computational complexity by selectively bypassing network layers, whereas frequency scaling adjusts the frequency of the processor to optimize energy use under latency constraints. Experiments of PLS and FS on ResNet-152 with the CIFAR-10 dataset demonstrated significant reductions in computational demands and energy consumption with minimal accuracy loss. This study offers practical solutions for improving real-time processing in resource-limited settings and provides insights into balancing computational efficiency and model performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_24014 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Optimization of Layer Skipping and Frequency Scaling for Convolutional Neural Networks under Latency Constraint Chan, Minh David Thao Zhao, Ruoyu Jia, Yukuan Mao, Ruiqing Zhou, Sheng Computer Vision and Pattern Recognition The energy consumption of Convolutional Neural Networks (CNNs) is a critical factor in deploying deep learning models on resource-limited equipment such as mobile devices and autonomous vehicles. We propose an approach involving Proportional Layer Skipping (PLS) and Frequency Scaling (FS). Layer skipping reduces computational complexity by selectively bypassing network layers, whereas frequency scaling adjusts the frequency of the processor to optimize energy use under latency constraints. Experiments of PLS and FS on ResNet-152 with the CIFAR-10 dataset demonstrated significant reductions in computational demands and energy consumption with minimal accuracy loss. This study offers practical solutions for improving real-time processing in resource-limited settings and provides insights into balancing computational efficiency and model performance. |
| title | Optimization of Layer Skipping and Frequency Scaling for Convolutional Neural Networks under Latency Constraint |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.24014 |