Tech

A promising photoelectron synapse for alpha-indium selenide-based reservoir computing


A promising photoelectron synapse for alpha-Indium Selenide-based reservoir computing

Multi-mode and multi-scale reservoir calculation. Credit: Liu et al.

Reservoir computing is an emerging computational framework based on the use of regressive neural networks (that is, networks in which data persists or recurs according to patterns). This framework has the potential to reduce data processing time, while improving the power efficiency of neural devices.

Researchers at Peking University and Beijing Academy of Artificial Intelligence recently introduced a new artificial synapse based on alpha-indium selenide (α-In)2se3 ), which may help to more efficiently replicate biological neural processes in neural devices. This synapse, introduced in a paper published in nature electronicscan have very valuable meanings for reservoir calculation applications.

Yuchao Yang, one of the researchers did: “Our idea arose from the need for a simple strategy that could be used to exploit the dynamic responses of physical systems to compute and physical reservoir computing is a promising framework for accomplishing this goal.” out of the study, told TechXplore.

“In2se3 is a very interesting material and a good foundation for reservoir computing, and its rich physical properties support the creation of multi-mode and multi-scale reservoir computing systems, which I hope to expand the application scenarios of physical reservoir computing.”

Reservoir computing relies on the use of artificial synapses that can run deep learning algorithms directly without having to transfer data between memory and the processing unit. Van der Waals α-In . Semiconductor Materials2se3 has many favorable optoelectronic, ferroelectric and semiconductor properties, making it an ideal candidate to fabricate these artificial synapses.

“In2se3 has two interesting physical properties simultaneously, namely ferroelectric switching and photoelectron response,” explains Yang. “We built a planar device to use in-plane ferroelectric polarizers for the electrical synapse, while also introducing light as the third terminal to trigger the reactive photoelectron. This unique structure effectively combines the two physical properties and can exploit the combination of ferroelectrics and optoelectronics to create heteroplasticity and higher-level computing functions.”

A promising photoelectron synapse for alpha-Indium Selenide-based reservoir computing

α-In2Se3 synapses with heterojunctional plasticity. Credit: Liu et al.

The temporal dynamics of the photoelectric synapse generated by this group of researchers can be controlled using electrical and optical stimuli. This means it could eventually artificially recreate the brain’s innate flexibility (i.e. its ability to adapt over time), while also processing information directly.

“The majority of previous studies in neurocomputing only used the device as a non-variable factor, while we took advantage of more complex non-linear dynamics to empower electricity,” said Yang. maths”.

“Compared with previous reservoir systems with fixed mechanisms and functions, our photoelectron synapse possesses both ferroelectric and optoelectronic properties, thus providing two mechanisms that combine the This allows us to realize a reservoir computing system based on mixed signal input, exhibiting excellent scalability and enhanced network performance. , while here multilevel signal processing is achieved by modulating the relaxation time of the device by either the tailgate voltage or the light.”

To evaluate the performance of the artificial synapse, Yang and his colleagues once built a multi-mode reservoir computing system. They then tested the system’s performance on a handwritten digit recognition task and a QR code recognition task. They found that it achieved promising results, successfully solving both of these data processing tasks with over 80% accuracy.

The artificial synapse realized as part of this research could soon open up exciting new possibilities for reservoir computing. In addition, the reservoir computing system created using this synapse can be further developed to solve other complex information processing and data analysis tasks.

“Our demonstration of a multi-mode and multi-scale reservoir computing system fundamentally expands the processing capabilities of reservoir Yang adds: “In our recent research, we focused on computing applications, but in the future we also want to realize a fully integrated neural simulation system, including perceiving information.”

More information:
Keqin Liu et al., An α-In2Se3-based optoelectronic synapse with controllable temporal dynamics for multi-mode and multi-scale reservoir computing, nature electronics (2022). DOI: 10.1038/s41928-022-00847-2

© 2022 Science X Network

Quote: A promising photoelectron synapse for alpha-indium selenide-based reservoir computing (2022, Nov 15) retrieved Nov 16, 2022 from https://techxplore.com/news/ 2022-11-optoelectronic-synapse-reservoir-based-alpha-indium .html

This document is the subject for the collection of authors. Other than any fair dealing for private learning or research purposes, no part may be reproduced without written permission. The content provided is for informational purposes only.

news7f

News7F: Update the world's latest breaking news online of the day, breaking news, politics, society today, international mainstream news .Updated news 24/7: Entertainment, Sports...at the World everyday world. Hot news, images, video clips that are updated quickly and reliably

Related Articles

Back to top button