Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2405.13672 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917675797577728 |
|---|---|
| author | Shan, Yimeng Zhang, Malu Zhu, Rui-jie Qiu, Xuerui Eshraghian, Jason K. Qu, Haicheng |
| author_facet | Shan, Yimeng Zhang, Malu Zhu, Rui-jie Qiu, Xuerui Eshraghian, Jason K. Qu, Haicheng |
| contents | Recent advancements in neuroscience research have propelled the development of Spiking Neural Networks (SNNs), which not only have the potential to further advance neuroscience research but also serve as an energy-efficient alternative to Artificial Neural Networks (ANNs) due to their spike-driven characteristics. However, previous studies often neglected the multiscale information and its spatiotemporal correlation between event data, leading SNN models to approximate each frame of input events as static images. We hypothesize that this oversimplification significantly contributes to the performance gap between SNNs and traditional ANNs. To address this issue, we have designed a Spiking Multiscale Attention (SMA) module that captures multiscale spatiotemporal interaction information. Furthermore, we developed a regularization method named Attention ZoneOut (AZO), which utilizes spatiotemporal attention weights to reduce the model's generalization error through pseudo-ensemble training. Our approach has achieved state-of-the-art results on mainstream neural morphology datasets. Additionally, we have reached a performance of 77.1% on the Imagenet-1K dataset using a 104-layer ResNet architecture enhanced with SMA and AZO. This achievement confirms the state-of-the-art performance of SNNs with non-transformer architectures and underscores the effectiveness of our method in bridging the performance gap between SNN models and traditional ANN models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_13672 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Advancing Spiking Neural Networks towards Multiscale Spatiotemporal Interaction Learning Shan, Yimeng Zhang, Malu Zhu, Rui-jie Qiu, Xuerui Eshraghian, Jason K. Qu, Haicheng Computer Vision and Pattern Recognition Recent advancements in neuroscience research have propelled the development of Spiking Neural Networks (SNNs), which not only have the potential to further advance neuroscience research but also serve as an energy-efficient alternative to Artificial Neural Networks (ANNs) due to their spike-driven characteristics. However, previous studies often neglected the multiscale information and its spatiotemporal correlation between event data, leading SNN models to approximate each frame of input events as static images. We hypothesize that this oversimplification significantly contributes to the performance gap between SNNs and traditional ANNs. To address this issue, we have designed a Spiking Multiscale Attention (SMA) module that captures multiscale spatiotemporal interaction information. Furthermore, we developed a regularization method named Attention ZoneOut (AZO), which utilizes spatiotemporal attention weights to reduce the model's generalization error through pseudo-ensemble training. Our approach has achieved state-of-the-art results on mainstream neural morphology datasets. Additionally, we have reached a performance of 77.1% on the Imagenet-1K dataset using a 104-layer ResNet architecture enhanced with SMA and AZO. This achievement confirms the state-of-the-art performance of SNNs with non-transformer architectures and underscores the effectiveness of our method in bridging the performance gap between SNN models and traditional ANN models. |
| title | Advancing Spiking Neural Networks towards Multiscale Spatiotemporal Interaction Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2405.13672 |