Modeling learnable electrical synapse for high precision spatio-temporal recognition

doi:10.1016/j.neunet.2022.02.006

CORC > 自动化研究所 > 中国科学院自动化研究所 > 数字内容技术与服务研究中心 > 听觉模型与认知计算

	Modeling learnable electrical synapse for high precision spatio-temporal recognition
	Wu, Zhenzhi 4; Zhang, Zhihong 3; Gao, Huanhuan 3; Qin, Jun 3; Zhao, Rongzhen 4; Zhao, Guangshe 3; Li, Guoqi 1,2
刊名	NEURAL NETWORKS
	2022-05-01
卷号	149 页码:184-194
关键词	Electrical synapse coupling Leaky-integrate-and-fire model Spatio-temporal information Bio-plausible neuronal dynamics
ISSN号	0893-6080
DOI	10.1016/j.neunet.2022.02.006
通讯作者	Li, Guoqi(guoqi.li@ia.ac.cn)
英文摘要	Bio-inspired recipes are being introduced to artificial neural networks for the efficient processing of spatio-temporal tasks. Among them, Leaky Integrate and Fire (LIF) model is the most remarkable one thanks to its temporal processing capability, lightweight model structure, and well investigated direct training methods. However, most learnable LIF networks generally take neurons as independent individuals that communicate via chemical synapses, leaving electrical synapses all behind. On the contrary, it has been well investigated in biological neural networks that the inter-neuron electrical synapse takes a great effect on the coordination and synchronization of generating action potentials. In this work, we are engaged in modeling such electrical synapses in artificial LIF neurons, where membrane potentials propagate to neighbor neurons via convolution operations, and the refined neural model ECLIF is proposed. We then build deep networks using ECLIF and trained them using a back-propagation-through-time algorithm. We found that the proposed network has great accuracy improvement over traditional LIF on five datasets and achieves high accuracy on them. In conclusion, it reveals that the introduction of the electrical synapse is an important factor for achieving high accuracy on realistic spatio-temporal tasks.
WOS关键词	SPIKING NEURONS ; FIRE MODEL ; BACKPROPAGATION ; MECHANISMS ; NETWORKS ; DYNAMICS
WOS研究方向	Computer Science ; Neurosciences & Neurology
语种	英语
出版者	PERGAMON-ELSEVIER SCIENCE LTD
WOS记录号	WOS:000793060100014
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/49401]
专题	数字内容技术与服务研究中心_听觉模型与认知计算
通讯作者	Li, Guoqi
作者单位	1.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Automaton, Beijing 100190, Peoples R China 3.Xi An Jiao Tong Univ, Sch Automation Sci & Engn, Xian 710049, Shaanxi, Peoples R China 4.Lynxi Technol, Beijing 100097, Peoples R China
推荐引用方式 GB/T 7714	Wu, Zhenzhi,Zhang, Zhihong,Gao, Huanhuan,et al. Modeling learnable electrical synapse for high precision spatio-temporal recognition[J]. NEURAL NETWORKS,2022,149:184-194.
APA	Wu, Zhenzhi.,Zhang, Zhihong.,Gao, Huanhuan.,Qin, Jun.,Zhao, Rongzhen.,...&Li, Guoqi.(2022).Modeling learnable electrical synapse for high precision spatio-temporal recognition.NEURAL NETWORKS,149,184-194.
MLA	Wu, Zhenzhi,et al."Modeling learnable electrical synapse for high precision spatio-temporal recognition".NEURAL NETWORKS 149(2022):184-194.