Understanding the Functional Roles of Modelling Components in Spiking Neural Networks
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_ | 1866929687331078144 |
|---|---|
| author | Yin, Huifeng Zheng, Hanle Mao, Jiayi Ding, Siyuan Liu, Xing Xu, Mingkun Hu, Yifan Pei, Jing Deng, Lei |
| author_facet | Yin, Huifeng Zheng, Hanle Mao, Jiayi Ding, Siyuan Liu, Xing Xu, Mingkun Hu, Yifan Pei, Jing Deng, Lei |
| contents | Spiking neural networks (SNNs), inspired by the neural circuits of the brain, are promising in achieving high computational efficiency with biological fidelity. Nevertheless, it is quite difficult to optimize SNNs because the functional roles of their modelling components remain unclear. By designing and evaluating several variants of the classic model, we systematically investigate the functional roles of key modelling components, leakage, reset, and recurrence, in leaky integrate-and-fire (LIF) based SNNs. Through extensive experiments, we demonstrate how these components influence the accuracy, generalization, and robustness of SNNs. Specifically, we find that the leakage plays a crucial role in balancing memory retention and robustness, the reset mechanism is essential for uninterrupted temporal processing and computational efficiency, and the recurrence enriches the capability to model complex dynamics at a cost of robustness degradation. With these interesting observations, we provide optimization suggestions for enhancing the performance of SNNs in different scenarios. This work deepens the understanding of how SNNs work, which offers valuable guidance for the development of more effective and robust neuromorphic models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_16674 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Understanding the Functional Roles of Modelling Components in Spiking Neural Networks Yin, Huifeng Zheng, Hanle Mao, Jiayi Ding, Siyuan Liu, Xing Xu, Mingkun Hu, Yifan Pei, Jing Deng, Lei Neural and Evolutionary Computing Artificial Intelligence Machine Learning Spiking neural networks (SNNs), inspired by the neural circuits of the brain, are promising in achieving high computational efficiency with biological fidelity. Nevertheless, it is quite difficult to optimize SNNs because the functional roles of their modelling components remain unclear. By designing and evaluating several variants of the classic model, we systematically investigate the functional roles of key modelling components, leakage, reset, and recurrence, in leaky integrate-and-fire (LIF) based SNNs. Through extensive experiments, we demonstrate how these components influence the accuracy, generalization, and robustness of SNNs. Specifically, we find that the leakage plays a crucial role in balancing memory retention and robustness, the reset mechanism is essential for uninterrupted temporal processing and computational efficiency, and the recurrence enriches the capability to model complex dynamics at a cost of robustness degradation. With these interesting observations, we provide optimization suggestions for enhancing the performance of SNNs in different scenarios. This work deepens the understanding of how SNNs work, which offers valuable guidance for the development of more effective and robust neuromorphic models. |
| title | Understanding the Functional Roles of Modelling Components in Spiking Neural Networks |
| topic | Neural and Evolutionary Computing Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2403.16674 |