Engineers Develop High-Performance, High-Reliability Artificial Synaptic Semiconductor Device

Engineers Develop High-Performance, High-Reliability Artificial Synaptic Semiconductor Device

Credit: Korea Institute of Science and Technology (KIST)

The technology of neuromorphic computing systems mimicking the human brain must overcome the limitation of excessive power consumption, which is characteristic of von Neumann’s existing computing method. A high-performance analog artificial synapse device capable of expressing the connection strength of synapses is needed to implement a semiconductor device that uses a brain information transmission method. This method uses signals transmitted between neurons when a neuron generates a spike signal.

However, with conventional variable resistance memory devices widely used as artificial synapses, as the filament grows with variable resistance, the electric field increases, causing a feedback phenomenon, causing the filament to grow rapidly. Therefore, it is difficult to implement plasticity while maintaining analog (gradual) resistance variation regarding filament type.

The Korea Institute of Science and Technology, led by Dr. YeonJoo Jeong’s team at the Center for Neuromorphic Engineering, has solved the limitations of analog synaptic characteristics, plasticity, and information preservation, which are chronic hurdles regarding memristors, neuromorphic semiconductor devices. He announced the development of an artificial synaptic semiconductor device capable of highly reliable neuromorphic computation.

The KIST research team refined the redox properties of active electrode ions to address small synaptic plasticity issues hampering the performance of existing neuromorphic semiconductor devices. Additionally, transition metals were doped and used in the synaptic device, controlling the probability of active electrode ion reduction. Engineers have discovered that the high probability of ion reduction is a critical variable in the development of high-performance artificial synaptic devices.

High-performance and high-reliability artificial synaptic semiconductor device for brain-mimicking next-generation computing

Example of visual information processing technology using the artificial synaptic device, confirming that the error rate is reduced by more than 60% by improving the performance of the device. Credit: Korea Institute of Science and Technology (KIST)

Therefore, a titanium transition metal, having a high probability of ion reduction, was introduced by the research team into an existing artificial synaptic device. This maintains the analog characteristics of the synapse and the plasticity of the device at the synapse of the biological brain, approximately five times the difference between high and low resistances. Additionally, they have developed a high-performance neuromorphic semiconductor that is about 50 times more efficient.

Moreover, due to the high alloying reaction exhibited by the doped titanium transition metal, the information retention increased up to 63 times compared to the existing artificial synaptic device. Additionally, brain functions, including long-term potentiation and long-term depression, could be more accurately simulated.

The team implemented an artificial neural network learning model using the developed artificial synaptic device and attempted artificial intelligence image recognition learning. The error rate was reduced by more than 60% compared to the existing artificial synaptic device; in addition, handwriting image pattern recognition (MNIST) accuracy increased by more than 69%. The research team confirmed the feasibility of a high-performance neuromorphic computing system through this improvement of the artificial synaptic device.

High-performance and high-reliability artificial synaptic semiconductor device for brain-mimicking next-generation computing

Photographs of (a) Solar energy collector, (b) Membrane distillation system. Credit: Korea Institute of Science and Technology (KIST)

Dr. Jeong of KIST said, “This study has significantly improved synaptic motion range and information preservation, which were the biggest technical hurdles of existing synaptic mimics. In the developed artificial synapse device, the operating area analog of the device to express the various connections of the synapse forces have been maximized, so that the performance of artificial intelligence based on brain simulation will be improved.

“In the follow-up research, we will fabricate a neuromorphic semiconductor chip based on the developed artificial synapse device to realize a high-performance artificial intelligence system, thus further enhancing the competitiveness in the home system and semiconductor field. of artificial intelligence.”

The research has been published in Nature Communication.


Neuromorphic memory device simulates neurons and synapses


More information:
Jaehyun Kang et al, Redox Dynamics Engineering Cluster-Type Analog Memristor for High-Performance Neuromorphic Computing, Nature Communication (2022). DOI: 10.1038/s41467-022-31804-4

Provided by the National Science and Technology Research Council

Quote: Engineers develop high-performance, high-reliability artificial synaptic semiconductor device (September 20, 2022) Retrieved September 20, 2022 from https://techxplore.com/news/2022-09-high-performance-high-reliability-artificial -synaptic-semiconductor.html

This document is subject to copyright. Except for fair use for purposes of private study or research, no part may be reproduced without written permission. The content is provided for information only.


#Engineers #Develop #HighPerformance #HighReliability #Artificial #Synaptic #Semiconductor #Device

Leave a Comment

Your email address will not be published.