A Novel Computing Paradigm for MobileNetV3 using Memristor
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910902125592576 |
|---|---|
| author | Li, Jiale Liu, Zhihang Ma, Sean Longyu Sham, Chiu-Wing Fu, Chong |
| author_facet | Li, Jiale Liu, Zhihang Ma, Sean Longyu Sham, Chiu-Wing Fu, Chong |
| contents | The increasing computational demands of deep learning models pose significant challenges for edge devices. To address this, we propose a memristor-based circuit design for MobileNetV3, specifically for image classification tasks. Our design leverages the low power consumption and high integration density of memristors, making it suitable for edge computing. The architecture includes optimized memristive convolutional modules, batch normalization modules, activation function modules, global average pooling modules, and fully connected modules. Experimental results on the CIFAR-10 dataset show that our memristor-based MobileNetV3 achieves over 90% accuracy while significantly reducing inference time and energy consumption compared to traditional implementations. This work demonstrates the potential of memristor-based designs for efficient deployment of deep learning models in resource-constrained environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_10512 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Novel Computing Paradigm for MobileNetV3 using Memristor Li, Jiale Liu, Zhihang Ma, Sean Longyu Sham, Chiu-Wing Fu, Chong Hardware Architecture The increasing computational demands of deep learning models pose significant challenges for edge devices. To address this, we propose a memristor-based circuit design for MobileNetV3, specifically for image classification tasks. Our design leverages the low power consumption and high integration density of memristors, making it suitable for edge computing. The architecture includes optimized memristive convolutional modules, batch normalization modules, activation function modules, global average pooling modules, and fully connected modules. Experimental results on the CIFAR-10 dataset show that our memristor-based MobileNetV3 achieves over 90% accuracy while significantly reducing inference time and energy consumption compared to traditional implementations. This work demonstrates the potential of memristor-based designs for efficient deployment of deep learning models in resource-constrained environments. |
| title | A Novel Computing Paradigm for MobileNetV3 using Memristor |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2402.10512 |