A framework for the general design and computation of hybrid neural networks.


Journal

Nature communications
ISSN: 2041-1723
Titre abrégé: Nat Commun
Pays: England
ID NLM: 101528555

Informations de publication

Date de publication:
14 06 2022
Historique:
received: 31 12 2021
accepted: 25 05 2022
entrez: 14 6 2022
pubmed: 15 6 2022
medline: 18 6 2022
Statut: epublish

Résumé

There is a growing trend to design hybrid neural networks (HNNs) by combining spiking neural networks and artificial neural networks to leverage the strengths of both. Here, we propose a framework for general design and computation of HNNs by introducing hybrid units (HUs) as a linkage interface. The framework not only integrates key features of these computing paradigms but also decouples them to improve flexibility and efficiency. HUs are designable and learnable to promote transmission and modulation of hybrid information flows in HNNs. Through three cases, we demonstrate that the framework can facilitate hybrid model design. The hybrid sensing network implements multi-pathway sensing, achieving high tracking accuracy and energy efficiency. The hybrid modulation network implements hierarchical information abstraction, enabling meta-continual learning of multiple tasks. The hybrid reasoning network performs multimodal reasoning in an interpretable, robust and parallel manner. This study advances cross-paradigm modeling for a broad range of intelligent tasks.

Identifiants

pubmed: 35701391
doi: 10.1038/s41467-022-30964-7
pii: 10.1038/s41467-022-30964-7
pmc: PMC9198039
doi:

Types de publication

Journal Article Research Support, Non-U.S. Gov't

Langues

eng

Sous-ensembles de citation

IM

Pagination

3427

Informations de copyright

© 2022. The Author(s).

Références

Neural Netw. 2019 Mar;111:47-63
pubmed: 30682710
Proc Mach Learn Res. 2017;70:3987-3995
pubmed: 31909397
Front Comput Neurosci. 2016 Sep 14;10:94
pubmed: 27683554
Nat Commun. 2016 Sep 26;7:12805
pubmed: 27666698
Nat Neurosci. 2019 Nov;22(11):1761-1770
pubmed: 31659335
Nature. 2020 Oct;586(7829):378-384
pubmed: 33057220
Neuron. 2017 Jul 19;95(2):245-258
pubmed: 28728020
Sensors (Basel). 2015 Nov 03;15(11):27894-904
pubmed: 26540060
Front Neurosci. 2021 Mar 05;15:633945
pubmed: 33746701
Proc Natl Acad Sci U S A. 2017 Mar 28;114(13):3521-3526
pubmed: 28292907
Mol Psychiatry. 2015 Feb;20(2):263-74
pubmed: 24514566
Nature. 2019 Nov;575(7784):607-617
pubmed: 31776490
J Neural Eng. 2007 Jun;4(2):L14-21
pubmed: 17409470
Nat Neurosci. 2015 Feb;18(2):310-5
pubmed: 25531570
Neural Comput. 2006 Feb;18(2):245-82
pubmed: 16378515
Front Neurosci. 2018 May 23;12:331
pubmed: 29875621
Nature. 2019 Aug;572(7767):106-111
pubmed: 31367028
Proc Natl Acad Sci U S A. 2018 Oct 30;115(44):E10467-E10475
pubmed: 30315147
Science. 2012 Nov 30;338(6111):1202-5
pubmed: 23197532

Auteurs

Rong Zhao (R)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.
IDG/McGovern Institute for Brain Research at Tsinghua University, 100084, Beijing, China.

Zheyu Yang (Z)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Hao Zheng (H)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Yujie Wu (Y)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Faqiang Liu (F)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Zhenzhi Wu (Z)

Lynxi Technologies Co., Ltd, 100080, Beijing, China.

Lukai Li (L)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Feng Chen (F)

Department of Automation, Tsinghua University, 100084, Beijing, China.

Seng Song (S)

Department of Biomedical Engineering, Tsinghua University, 100084, Beijing, China.

Jun Zhu (J)

Department of Computer Science and Technology, Tsinghua University, 100084, Beijing, China.

Wenli Zhang (W)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Haoyu Huang (H)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Mingkun Xu (M)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Kaifeng Sheng (K)

Lynxi Technologies Co., Ltd, 100080, Beijing, China.

Qianbo Yin (Q)

Lynxi Technologies Co., Ltd, 100080, Beijing, China.

Jing Pei (J)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Guoqi Li (G)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China.

Youhui Zhang (Y)

Department of Computer Science and Technology, Tsinghua University, 100084, Beijing, China.

Mingguo Zhao (M)

Department of Automation, Tsinghua University, 100084, Beijing, China.

Luping Shi (L)

Center for Brain-Inspired Computing Research (CBICR), Beijing Advanced Innovation Center for Integrated Circuits, Optical Memory National Engineering Research Center, & Department of Precision Instrument, Tsinghua University, 100084, Beijing, China. lpshi@tsinghua.edu.cn.
IDG/McGovern Institute for Brain Research at Tsinghua University, 100084, Beijing, China. lpshi@tsinghua.edu.cn.

Articles similaires

alpha-Synuclein Humans Animals Mice Lewy Body Disease

Unsupervised learning for real-time and continuous gait phase detection.

Dollaporn Anopas, Yodchanan Wongsawat, Jetsada Arnin
1.00
Humans Gait Neural Networks, Computer Unsupervised Machine Learning Walking
Animals Optogenetics Visual Cortex Neurons Mice
Humans Shoulder Fractures Tomography, X-Ray Computed Neural Networks, Computer Female

Classifications MeSH