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

Rong Zhao, Zheyu Yang, Hao Zheng, Yujie Wu, Faqiang Liu, Zhenzhi Wu, Lukai Li, Feng Chen, Seng Song, Jun Zhu, Wenli Zhang, Haoyu Huang, Mingkun Xu, Kaifeng Sheng, Qianbo Yin, Jing Pei, Guoqi Li, Youhui Zhang, Mingguo Zhao and Luping Shi ()
Additional contact information
Rong Zhao: Tsinghua University
Zheyu Yang: Tsinghua University
Hao Zheng: Tsinghua University
Yujie Wu: Tsinghua University
Faqiang Liu: Tsinghua University
Zhenzhi Wu: Lynxi Technologies Co., Ltd
Lukai Li: Tsinghua University
Feng Chen: Tsinghua University
Seng Song: Tsinghua University
Jun Zhu: Tsinghua University
Wenli Zhang: Tsinghua University
Haoyu Huang: Tsinghua University
Mingkun Xu: Tsinghua University
Kaifeng Sheng: Lynxi Technologies Co., Ltd
Qianbo Yin: Lynxi Technologies Co., Ltd
Jing Pei: Tsinghua University
Guoqi Li: Tsinghua University
Youhui Zhang: Tsinghua University
Mingguo Zhao: Tsinghua University
Luping Shi: Tsinghua University

Nature Communications, 2022, vol. 13, issue 1, 1-12

Abstract: Abstract 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.

Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
https://www.nature.com/articles/s41467-022-30964-7 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:13:y:2022:i:1:d:10.1038_s41467-022-30964-7

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-022-30964-7

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().