10 x microsfts, page-2

Trying to get a real answer from the BRN team was a massive mistake, they think i am out to ruin their families and existence

this is the paper, is BRN working on this exact tech,

Cheers

Weebit Nano & Polimi present paper on novel AI selflearning ReRAM hardware at leading industry
conference
21 July 2020 – Weebit Nano (ASX: WBT), the next generation memory technology for the global semiconductor
industry, and the Politecnico di Milano in Italy (Polimi), a leading European university for Industrial and
Information Engineering, Technology and Industrial Design, have presented a joint research paper on a novel
Artificial Intelligence (AI) self-learning demonstration based on Weebit’s silicon oxide (SiOx) ReRAM at a
prominent industry conference.
Held virtually in June, the VLSI Technology and Circuits Symposia is the premier conference for the international
semiconductor and circuits industry.
Presented by Polimi at the Symposia on VLSI Technology and Circuits, the paper outlines a brain-inspired AI
system which can perform unsupervised learning tasks with high accuracy results.
It uses Polimi’s developed hardware design and Weebit’s silicon oxide ReRAM to combine the efficiency of the
state of the art Convolutional Neural Networks (CNN) with the plasticity of brain-inspired Spiking Neural
Networks, enabling the hardware to learn new things without forgetting trained tasks of previously acquired
information. In this way, it enables unsupervised learning typical of the human brain which learns new skills
throughout its whole life adapting to its environment, without forgetting older information.
The system’s accuracy has been validated by standard databases, including digit recognition (MNIST) at 99.3
percent, dataset of clothing (Fashion-MNIST) at 93 percent and CIFAR-10 (various object recognition dataset) at
91 percent.
In addition, the demonstration adapts its operative frequency for power saving, enabling continual learning of up
to 50 per cent for non-trained classes. The use of power saving spike-frequency modulation enables feasible
solutions for lifelong learning in autonomous AI systems.
Professor Ielmini, recipient of the Intel 2013 Outstanding Researcher Award, who has held visiting positions at
Intel and Stanford University and conducted research on emerging non-volatile memories for several years, said:
“Continual learning is essential for us as humans to accumulate knowledge. Artificial neural networks currently
lack this ability, as the previous knowledge is generally erased by a second training – a process known as
catastrophic forgetting. This AI system combines the best of both worlds, namely the accuracy of deep learning
and the flexibility of the human brain, thus moving one step closer to the realisation of brain-like hardware.”
Coby Hanoch, CEO of Weebit Nano, said: “Weebit’s progress with Professor Ielmini on a joint neuromorphic
ReRAM project over the past year, demonstrates the capability of our silicon oxide ReRAM technology in artificial
intelligence applications. Our ongoing collaboration with Polimi will ensure our technology is at the forefront of
future artificial intelligence and neuromorphic computing applications, addressing the challenges of tomorrow.