A new technical paper titled “An Artificial Neuron with a Leaky Fin-Shaped Field-Effect Transistor for a Highly Scalable Capacitive Neural Network” was published by researchers at KAIST (Korea Advanced Institute of Science and Technology).
“In commercialized flash memory, tunnelling oxide prevents the trapped charges from escaping for better memory ability. In our proposed FinFET neuron, tunnelling oxide was intentionally removed for escaping of the trapped charges giving us the leaky function of the neuron,” said the paper’s lead author and KAIST researcher Joon-Kyu Han in this Advanced Science News article. “Thus the leaky integrate-and-fire (LIF) function of the biological neuron was mimicked thanks to the gate structure of the proposed FinFET neuron.”
Find the technical paper here. Published October 2022.
Han, J., Yu, J., Kim, D. and Choi, Y. (2022), An Artificial Neuron with a Leaky Fin-Shaped Field-Effect Transistor for a Highly Scalable Capacitive Neural Network. Adv. Intell. Syst. 2200112. https://doi.org/10.1002/aisy.202200112.
Related Reading
Neuromorphic Computing Knowledge Center
Developers Turn To Analog For Neural Nets
Replacing digital with analog circuits and photonics can improve performance and power, but it’s not that simple.
How To Optimize A Processor
There are at least three architectural layers to processor design, each of which plays a significant role.
Complex Tradeoffs In Inferencing Chips
One size does not fit all, and each processor type has its own benefits and drawbacks.
Leave a Reply