Artificial Intelligence Inspires Physicists to Rework Hardware
Physicists from Purdue University, University of California San Diego (USCD), and École Supérieure de Physique et de Chimie Industrielles (ESPCI) are on a mission to rework hardware to handle the complexity of tomorrow’s computational advances by mimicking the synapses of the human brain. Their research suggests that neuromorphic architectures hold promise for lower energy consumption processors, enhanced computation, fundamentally different computational modes, native learning, and enhanced pattern recognition, all of which are necessary to handle the workload of new AI technological breakthroughs.
What is Neuromorphic Architecture?
Neuromorphic architecture is all about computer chips mimicking brain behavior by using neurons and synapses to transmit information and encode memory. The team of scientists discovered that vanadium oxides show tremendous promise for neuromorphic computing because they can be used to make both artificial neurons and synapses.
The Promise of Vanadium Oxides
The team’s research led to the discovery of vanadium oxides as possible future neuromorphic devices. According to the lead experimental scientist, Alexandre Zimmers, the unique properties of vanadium oxides allow for the accumulation of memory throughout the entirety of the sample, rather than just at the boundaries of domains, opening new opportunities for control.
The team plans to continue their research to improve the synaptic behavior of vanadium oxides and enhance their potential for future neuromorphic devices. By locally tweaking and observing the effects of ion bombardment on the material’s surface, they hope to guide the electrical current through specific regions in the sample where the memory effect is at its maximum. This has the potential to significantly enhance the synaptic behavior of this neuromorphic material.