Abstract

Although the energy consumption of reported neuromorphic computing devices inspired by biological systems has become lower than traditional memory, it still remains greater than bio‐synapses (≈10 fJ per spike). Herein, a flexible MoS2‐based heterosynapse is designed with two modulation modes, an electronic mode and a photoexcited mode. A one‐step mechanical exfoliation method on flexible substrate and low‐temperature atomic layer deposition process compatible with flexible electronics are developed for fabricating wearable heterosynapses. With a pre‐spike of 100 ns, the synaptic device exhibits ultralow energy consumption of 18.3 aJ per spike in long‐term potentiation and 28.9 aJ per spike in long‐term depression. The ultrafast speed and ultralow power consumption provide a path for a neuromorphic computing system owning more excellent processing ability than the human brain. By adding optical modulation, a modulatory synapse is constructed to dynamically control correlations between pre‐ and post‐synapses and realize complex global neuromodulations. The novel wearable heterosynapse expands the accessible range of synaptic weights (ratio of facilitation ≈228%), providing an insight into the application of wearable 2D highly efficient neuromorphic computing architectures.

Highlights

  • The energy consumption of reported neuromorphic computing which consume 10 fJ per synaptic event

  • Most previous reports were limited to single devices with simple connections and pre- and post-synaptic learning rules, such as excitatory post-synaptic current (EPSC), inhibitory post-synaptic current architectures

  • (IPSC), short-term plasticity (STP), paired pulse facilitation/depression (PPF/PPD), spike time-dependent plasticity (STDP), Conventional von Neumann architectures based on comple- learning-forgetting-relearning behaviors etc.[11,12,13]

Read more

Summary

Introduction

The energy consumption of reported neuromorphic computing which consume 10 fJ per synaptic event. We demonstrate ultralow energy consumption (

Results
Conclusion
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.