First mover advantage, page-39

The following extract, link to the full paper and the Acknowledgments as to who is funding this work is very interesting given that NASA & DARPA are clearly aware of the only COTS AKD1000 Spiking Neural Network processor on the market and already in testing in their labs and being implemented by an established DARPA defence contractor:

“Another approach researchers have taken to mitigate the von Neumann bottleneck is to safely parallelizing data flow between the CPU and memory – like how the brain accesses information using multiple neurons at the same time without any interference.[438] By parallelizing the CPU’s data access, one can transfer more information (using the same transfer rate per bandwidth) in the same amount of time. Recent studies have investigated this neuromorphic model using artificial intelligence to create non-von Neumann architectures.[439] One such promising artificial intelligent model is the Spiking Neural Network (SNN) design. In SNNs, data is only sent when a given node receives an input voltage (a spike) above some threshold, allowing multiple nodes to be activated at the same time.[439] In literature, SNNs have been used in dynamic neuromorphic asynchronous This article is protected by copyright. All rights reserved. 41 processors with an insignificant time delay, outperforming other modern von Neumann architectures in accessing stored data.[439] Further research into the integration of ML in computer architectures can enhance data utilization and parallelization, fundamentally changing the speed at which computers transmit information. Pages 40 -41”

https://authors.library.caltech.edu/112380/1/adma.202107902.pdf

Acknowledgements Wenzheng Heng and Samuel Solomon contributed equally to this work. This project was supported by the National Institutes of Health grant R01HL155815, Office of Naval Research grant N00014-21- 1-2483 and N00014-21-1-2845, the Translational Research Institute for Space Health through NASA NNX16AO69A, the Tobacco-Related Disease Research Program grant R01RG3746, and Carver Mead New Adventures Fund at California Institute of Technology. Conflict of Interest The authors declare no conflict of interest.

My opinion only DYOR
FF

AKIDA BALLISTA