2023-11-08 14:38:08
Certainly, technological advances in artificial intelligence (AI) bring ultra-fast computing closer thanks to the neuromorphic computer. But the question is whether the current infrastructure is ready to handle this workload.
In reality, the codes that humans write are often run on conventional silicon architectures. However, these latter are not suitable for this task. Researchers from Purdue University, the University of California at San Diego (USCD) and the École Supérieure de Physique et de Chimie Industrielles (ESPCI) in Paris then sought to resolve this challenge. They published their results in Advanced Electronic Materials, where they explored an approach to reshape the hardware by mimicking the synapses of the human brain.
A promising future for the neuromorphic computer
The neuromorphic computer, which mimics the behavior of the brain, relies on special computer chips. In the brain, neurons transmit information via synapses, which play a key role in memory. Researchers have found that vanadium oxides hold promise for neuromorphic computing because they allow the creation of both artificial neurons and synapses.
The architecture of neuromorphic computers has a major advantage: lower energy consumption than traditional silicon architectures. This is due to their ability to mimic the basic components of a brain, namely neurons and synapses. Unlike silicon, which is effective for memory storage, neuromorphic materials mimic neuronal behavior.
However, finding suitable materials to create both good synapses and good artificial neurons is a challenge. Only a few quantum materials show promise in this area, notably vanadium dioxide. The researchers found that memory accumulates throughout the vanadium sample. This discovery then opens up new possibilities for the control of this property.
The results of this research have been revealed. Microscopic videos, however, have shown that changes in the metallic and insulating domains of vanadium cause a memory accumulation in the entire sample.
This memory results from local temperature changes as the material transitions from insulator to metal and vice versa. There preferential diffusion of point defects in metallic domains seems to contribute to this accumulation of memory.
The researchers now plan to continue their work by locally modifying the vanadium and observing the effects. In particular the impact of the ion bombardment on the surface of the material.
This might allowimprove the synaptic behavior of this neuromorphic material. This is done by guiding the electric current towards the areas where the memory effect is most pronounced.
1699456895
#researchers #discovered #hardware #mimics #behavior #human #brain
