Spike-based dynamic computing with asynchronous sensing-computing neuromorphic chip

Man Yao, Ole Richter, Guangshe Zhao, Ning Qiao, Yannan Xing, Dingheng Wang, Tianxiang Hu, Wei Fang, Tugba Demirci, Michele Marchi, Lei Deng, Tianyi Yan, Carsten Nielsen, Sadique Sheik, Chenxi Wu, Yonghong Tian, Bo Xu and Guoqi Li ()
Additional contact information
Man Yao: Chinese Academy of Sciences
Ole Richter: SynSense AG Corporation
Guangshe Zhao: Xi’an Jiaotong University
Ning Qiao: SynSense AG Corporation
Yannan Xing: SynSense Corporation
Dingheng Wang: Northwest Institute of Mechanical & Electrical Engineering
Tianxiang Hu: Chinese Academy of Sciences
Wei Fang: Peking University
Tugba Demirci: SynSense AG Corporation
Michele Marchi: SynSense AG Corporation
Lei Deng: Tsinghua University
Tianyi Yan: Beijing Institute of Technology
Carsten Nielsen: SynSense AG Corporation
Sadique Sheik: SynSense AG Corporation
Chenxi Wu: SynSense AG Corporation
Yonghong Tian: Peking University
Bo Xu: Chinese Academy of Sciences
Guoqi Li: Chinese Academy of Sciences

Nature Communications, 2024, vol. 15, issue 1, 1-18

Abstract: Abstract By mimicking the neurons and synapses of the human brain and employing spiking neural networks on neuromorphic chips, neuromorphic computing offers a promising energy-efficient machine intelligence. How to borrow high-level brain dynamic mechanisms to help neuromorphic computing achieve energy advantages is a fundamental issue. This work presents an application-oriented algorithm-software-hardware co-designed neuromorphic system for this issue. First, we design and fabricate an asynchronous chip called “Speck”, a sensing-computing neuromorphic system on chip. With the low processor resting power of 0.42mW, Speck can satisfy the hardware requirements of dynamic computing: no-input consumes no energy. Second, we uncover the “dynamic imbalance” in spiking neural networks and develop an attention-based framework for achieving the algorithmic requirements of dynamic computing: varied inputs consume energy with large variance. Together, we demonstrate a neuromorphic system with real-time power as low as 0.70mW. This work exhibits the promising potentials of neuromorphic computing with its asynchronous event-driven, sparse, and dynamic nature.

Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-024-47811-6 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:15:y:2024:i:1:d:10.1038_s41467-024-47811-6

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

DOI: 10.1038/s41467-024-47811-6

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 ().