Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks
Summary: The spiking neural network (SNN) mimics the information-processing operation in the human brain. Directly applying backpropagation to the training of the SNN still has a performance gap compared with traditional deep neural networks. To address the problem, we propose a biologically plausib...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-06-01
|
| Series: | Patterns |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666389922001192 |
| _version_ | 1817975429767626752 |
|---|---|
| author | Guobin Shen Dongcheng Zhao Yi Zeng |
| author_facet | Guobin Shen Dongcheng Zhao Yi Zeng |
| author_sort | Guobin Shen |
| collection | DOAJ |
| description | Summary: The spiking neural network (SNN) mimics the information-processing operation in the human brain. Directly applying backpropagation to the training of the SNN still has a performance gap compared with traditional deep neural networks. To address the problem, we propose a biologically plausible spatial adjustment that rethinks the relationship between membrane potential and spikes and realizes a reasonable adjustment of gradients to different time steps. It precisely controls the backpropagation of the error along the spatial dimension. Secondly, we propose a biologically plausible temporal adjustment to make the error propagate across the spikes in the temporal dimension, which overcomes the problem of the temporal dependency within a single spike period of traditional spiking neurons. We have verified our algorithm on several datasets, and the experimental results have shown that our algorithm greatly reduces network latency and energy consumption while also improving network performance. The bigger picture: The spiking neural network (SNN) captures more important aspects of brain information processing and has been applied to various domains. The biggest problem restricting the development of SNN is the training algorithm. Backpropagation (BP)-based training has extended SNNs to more complex network structures and datasets. However, the traditional design of BP ignores the dynamic characteristics of SNNs and is not biologically plausible. This paper rethinks the problems in BP-based SNNs and proposes a biologically plausible spatiotemporal adjustment to replace the traditional artificial design. The adjustment greatly improves the performance of the SNNs and reduces energy consumption and latency. The long-term ambition of this research is to take more inspiration on learning mechanisms and structures from the cognitive brain at different levels of details to build even more biologically plausible SNNs as a foundation for future artificial intelligence models. |
| first_indexed | 2024-04-13T21:50:00Z |
| format | Article |
| id | doaj.art-63199c6dfaed4646ba18986f7f2bb0e0 |
| institution | Directory Open Access Journal |
| issn | 2666-3899 |
| language | English |
| last_indexed | 2024-04-13T21:50:00Z |
| publishDate | 2022-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Patterns |
| spelling | doaj.art-63199c6dfaed4646ba18986f7f2bb0e02022-12-22T02:28:27ZengElsevierPatterns2666-38992022-06-0136100522Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networksGuobin Shen0Dongcheng Zhao1Yi Zeng2Research Center for Brain-Inspired Intelligence, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China; School of Future Technology, University of Chinese Academy of Sciences, Beijing 100190, ChinaResearch Center for Brain-Inspired Intelligence, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, ChinaResearch Center for Brain-Inspired Intelligence, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China; Center for Excellence in Brain Science and Intelligence Technology, Chinese Academy of Sciences, Shanghai 200031, China; National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China; School of Future Technology, University of Chinese Academy of Sciences, Beijing 100190, China; School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100190, China; Corresponding authorSummary: The spiking neural network (SNN) mimics the information-processing operation in the human brain. Directly applying backpropagation to the training of the SNN still has a performance gap compared with traditional deep neural networks. To address the problem, we propose a biologically plausible spatial adjustment that rethinks the relationship between membrane potential and spikes and realizes a reasonable adjustment of gradients to different time steps. It precisely controls the backpropagation of the error along the spatial dimension. Secondly, we propose a biologically plausible temporal adjustment to make the error propagate across the spikes in the temporal dimension, which overcomes the problem of the temporal dependency within a single spike period of traditional spiking neurons. We have verified our algorithm on several datasets, and the experimental results have shown that our algorithm greatly reduces network latency and energy consumption while also improving network performance. The bigger picture: The spiking neural network (SNN) captures more important aspects of brain information processing and has been applied to various domains. The biggest problem restricting the development of SNN is the training algorithm. Backpropagation (BP)-based training has extended SNNs to more complex network structures and datasets. However, the traditional design of BP ignores the dynamic characteristics of SNNs and is not biologically plausible. This paper rethinks the problems in BP-based SNNs and proposes a biologically plausible spatiotemporal adjustment to replace the traditional artificial design. The adjustment greatly improves the performance of the SNNs and reduces energy consumption and latency. The long-term ambition of this research is to take more inspiration on learning mechanisms and structures from the cognitive brain at different levels of details to build even more biologically plausible SNNs as a foundation for future artificial intelligence models.http://www.sciencedirect.com/science/article/pii/S2666389922001192DSML2: Proof-of-concept: Data science output has been formulated, implemented, and tested for one domain/problem |
| spellingShingle | Guobin Shen Dongcheng Zhao Yi Zeng Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks Patterns DSML2: Proof-of-concept: Data science output has been formulated, implemented, and tested for one domain/problem |
| title | Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks |
| title_full | Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks |
| title_fullStr | Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks |
| title_full_unstemmed | Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks |
| title_short | Backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks |
| title_sort | backpropagation with biologically plausible spatiotemporal adjustment for training deep spiking neural networks |
| topic | DSML2: Proof-of-concept: Data science output has been formulated, implemented, and tested for one domain/problem |
| url | http://www.sciencedirect.com/science/article/pii/S2666389922001192 |
| work_keys_str_mv | AT guobinshen backpropagationwithbiologicallyplausiblespatiotemporaladjustmentfortrainingdeepspikingneuralnetworks AT dongchengzhao backpropagationwithbiologicallyplausiblespatiotemporaladjustmentfortrainingdeepspikingneuralnetworks AT yizeng backpropagationwithbiologicallyplausiblespatiotemporaladjustmentfortrainingdeepspikingneuralnetworks |