Samsung Announces Innovative Brain-On-Chip Technology

Samsung Electronics, a world leader in advanced semiconductor technology, announces a new idea that advances the world closer to developing neuromorphic chips that more closely resemble the human brain.

“Our idea is audacious, but pursuing it will push the boundaries of machine intelligence, neuroscience, and semiconductor technology,” said Dr Ham.

The paper titled ‘Neuromorphic electronics based on copying and pasting the brain’, by Nature Electronics proposes a method for copying the brain's neuronal connection map using a ground-breaking nanoelectrode array and pasting this map onto a high-density three-dimensional (3D) network of solid-state memories. It was developed by Donhee Ham, a research fellow of Samsung Advanced Institute of Technology (SAIT) and a professor at Harvard University, and Hongkun Park, a professor at Harvard University. 

According to experts, the brain's functions are determined by the wiring map of a huge number of neurons. The article proposes a method for returning to the original neuromorphic goal of brain reverse engineering, which was to replicate the structure and function of neural networks on a silicon chip.

As per the researchers, due to the nanoelectrode array's capacity to successfully access a large number of neurons, it can record their electrical impulses with high sensitivity. These massively parallel intracellular recordings provide information about the neural wiring map, identifying the locations and strength of connections between neurons. As a result of these telltale recordings, the neural wiring map can be retrieved or reproduced. The replicated neural map can then be pasted onto a network of non-volatile memories, such as commercial flash memory found in solid-state drives (SSDs), or onto novel memories, such as resistive random access memories (RRAM).

Additionally, the article provides a technique for swiftly pasting the neural wiring map into a memory network.
For further information, refer to this article.