NOVA: NoC-based Vector Unit for Mapping Attention Layers on a CNN Accelerator
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929336945213440 |
|---|---|
| author | Upadhyay, Mohit Juneja, Rohan Wong, Weng-Fai Peh, Li-Shiuan |
| author_facet | Upadhyay, Mohit Juneja, Rohan Wong, Weng-Fai Peh, Li-Shiuan |
| contents | Attention mechanisms are becoming increasingly popular, being used in neural network models in multiple domains such as natural language processing (NLP) and vision applications, especially at the edge. However, attention layers are difficult to map onto existing neuro accelerators since they have a much higher density of non-linear operations, which lead to inefficient utilization of today's vector units. This work introduces NOVA, a NoC-based Vector Unit that can perform non-linear operations within the NoC of the accelerators, and can be overlaid onto existing neuro accelerators to map attention layers at the edge. Our results show that the NOVA architecture is up to 37.8x more power-efficient than state-of-the-art hardware approximators when running existing attention-based neural networks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_04206 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | NOVA: NoC-based Vector Unit for Mapping Attention Layers on a CNN Accelerator Upadhyay, Mohit Juneja, Rohan Wong, Weng-Fai Peh, Li-Shiuan Hardware Architecture Artificial Intelligence Machine Learning B.2.4 Attention mechanisms are becoming increasingly popular, being used in neural network models in multiple domains such as natural language processing (NLP) and vision applications, especially at the edge. However, attention layers are difficult to map onto existing neuro accelerators since they have a much higher density of non-linear operations, which lead to inefficient utilization of today's vector units. This work introduces NOVA, a NoC-based Vector Unit that can perform non-linear operations within the NoC of the accelerators, and can be overlaid onto existing neuro accelerators to map attention layers at the edge. Our results show that the NOVA architecture is up to 37.8x more power-efficient than state-of-the-art hardware approximators when running existing attention-based neural networks. |
| title | NOVA: NoC-based Vector Unit for Mapping Attention Layers on a CNN Accelerator |
| topic | Hardware Architecture Artificial Intelligence Machine Learning B.2.4 |
| url | https://arxiv.org/abs/2405.04206 |