Samsung to “copy paste” the brain onto 3D chips – a precursor to human-like AI?
- The tech giant has proposed a method that would “copy and paste” a brain’s neuron wiring map onto 3D neuromorphic chips
- The move could serve as a ‘shortcut’ to AI systems that behave like real human brains
- There’s a glaring problem with complexity as Samsung’s vision is ‘highly ambitious’
Samsung may have cracked the door to human-like artificial intelligence open just a little, with its vision to develop brain-like chips, in a paper co-authored with Harvard University researchers. The essence of the vision put forward by the authors is best summed up by the two words, ‘copy’ and ‘paste’. By borrowing existing brain structures, the tech firm has proposed a method that would “copy and paste” a brain’s neuron wiring map to 3D neuromorphic chips.
Authored by engineers and scholars from Samsung and Harvard University, the paper titled Neuromorphic electronics based on copying and pasting the brain, suggests a way to copy the brain’s neuronal connection map. Basically it uses a nanoelectrode array developed by Harvard University professors Ham Don-hee and Park Hong-kun to paste the map onto a high-density three-dimensional network of solid-state memory chips, the technology for which Samsung has been a world leader.
What’s the 3D chips vision all about?
According to Samsung, using this copy and paste approach the authors envision creating a memory chip that approximates the unique computing traits of the brain – low power, facile learning, adaptation to the environment, even autonomy and cognition – that have been beyond the reach of current technology.
The original goal of neuromorphic engineering was launched in the 1980s, which was to mimic the structure and function of the neuronal networks on a silicon chip. It proved difficult because, even until now, little is known of how a large number of neurons are wired together to create the brain’s higher functions.
Thus, the goal of neuromorphic engineering has been to design a chip ‘inspired’ by the brain rather than rigorously mimicking it. Also, the knowledge of the map is the key to reverse-engineering the brain since the brain is made up of a large number of neurons, and their wiring map is responsible for the brain’s functions.
The paper then suggests a way to return to the original neuromorphic goal of the brain reverse engineering. “The nanoelectrode array can effectively enter a large number of neurons so it can record their electrical signals with high sensitivity. These massively parallel intracellular recordings inform the neuronal wiring map, indicating where neurons connect with one another and how strong these connections are. Hence from these telltale recordings, the neuronal wiring map can be extracted, or ‘copied’,” it noted.
The copied neuronal map can then be ‘pasted’ to a network of non-volatile memories – such as commercial flash memories that are used in our everyday life in solid-state drives (SSD), or ‘new’ memories such as resistive random access memories (RRAM) – by programming each memory so that its conductance represents the strength of each neuronal connection in the copied map.
The hurdle to mimic the structure of the human brain
Given the human brain has an estimated 100 billion or so neurons, and a thousand or so times more synaptic connections, the ultimate neuromorphic chip will require 100 trillion or so memories. The authors reckon integrating such a vast number of memories on a single chip would be made possible by 3D integration of memories, the technology led by Samsung that opened up a new era for the memory industry.
Leveraging its leading experience in chip manufacturing, Samsung is planning to continue its research into neuromorphic engineering, in order to extend Samsung’s leadership in the field of next-generation AI semiconductors. “The vision we present is highly ambitious,” said Ham. “But working toward such a heroic goal will push the boundaries of machine intelligence, neuroscience, and semiconductor technology.”
22 February 2024
21 February 2024
21 February 2024