Rethinking Pretraining as a Bridge From ANNs to SNNs
文献类型：期刊论文
| 作者 | Lin, Yihan7; Hu, Yifan7; Ma, Shijie5,6 ; Yu, Dongjie4; Li, Guoqi1,2,3
|
| 刊名 | IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
 |
| 出版日期 | 2022-11-14 |
| 页码 | 14 |
| 关键词 | Training Task analysis Neurons Pipelines Artificial neural networks Feature extraction Transfer learning Event-driven dataset neural network (NN) analysis pretraining technique spiking NN (SNN) transfer learning |
| ISSN号 | 2162-237X |
| DOI | 10.1109/TNNLS.2022.3217796 |
| 通讯作者 | Li, Guoqi(guoqi.li@ia.ac.cn) |
| 英文摘要 | Spiking neural networks (SNNs) are known as typical kinds of brain-inspired models with their unique features of rich neuronal dynamics, diverse coding schemes, and low power consumption properties. How to obtain a high-accuracy model has always been the main challenge in the field of SNN. Currently, there are two mainstream methods, i.e., obtaining a converted SNN through converting a well-trained artificial NN (ANN) to its SNN counterpart or training an SNN directly. However, the inference time of a converted SNN is too long, while SNN training is generally very costly and inefficient. In this work, a new SNN training paradigm is proposed by combining the concepts of the two different training methods with the help of the pretrain technique and BP-based deep SNN training mechanism. We believe that the proposed paradigm is a more efficient pipeline for training SNNs. The pipeline includes pipe-S for static data transfer tasks and pipe-D for dynamic data transfer tasks. State-of-the-art (SOTA) results are obtained in a large-scale event-driven dataset ES-ImageNet. For training acceleration, we achieve the same (or higher) best accuracy as similar leaky-integrate-and-fire (LIF)-SNNs using 1/8 training time on ImageNet-1K and 1/2 training time on ES-ImageNet and also provide a time-accuracy benchmark for a new dataset ES-UCF101. These experimental results reveal the similarity of the functions of parameters between ANNs and SNNs and also demonstrate various potential applications of this SNN training pipeline. |
| WOS关键词 | INTELLIGENCE ; NETWORKS |
| 资助项目 | National Natural Science Foundation of China[61836004] ; National Natural Science Foundation of China[62236009] ; National Natural Science Foundation of China[U22A20103] ; Beijing Natural Science Foundation[JQ21015] ; National Key Research and Development Program of China[2018AAA0102600] ; Beijing Academy of Artificial Intelligence |
| WOS研究方向 | Computer Science ; Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:000888970400001 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 资助机构 | National Natural Science Foundation of China ; Beijing Natural Science Foundation ; National Key Research and Development Program of China ; Beijing Academy of Artificial Intelligence |
| 源URL | [http://ir.ia.ac.cn/handle/173211/51284]  |
| 专题 | 自动化研究所_模式识别国家重点实验室_模式分析与学习团队 |
| 通讯作者 | Li, Guoqi |
| 作者单位 | 1.Peng Cheng Lab, Shenzhen 518055, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 101408, Peoples R China 3.Chinese Acad Sci, Inst Automat, Beijing 100045, Peoples R China 4.Tsinghua Univ, Sch Vehicle & Mobil, Beijing 100084, Peoples R China 5.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 6.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 7.Tsinghua Univ, Dept Precis Instrument, Beijing 100084, Peoples R China |
| 推荐引用方式 GB/T 7714 | Lin, Yihan,Hu, Yifan,Ma, Shijie,et al. Rethinking Pretraining as a Bridge From ANNs to SNNs[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2022:14. |
| APA | Lin, Yihan,Hu, Yifan,Ma, Shijie,Yu, Dongjie,&Li, Guoqi.(2022).Rethinking Pretraining as a Bridge From ANNs to SNNs.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,14. |
| MLA | Lin, Yihan,et al."Rethinking Pretraining as a Bridge From ANNs to SNNs".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2022):14. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。