The electronic chip that can be reprogrammed on demand, which could help create artificial intelligence that learns continuously like the human brain. Image: Purdue University/Rebecca McElhoe
The researchers used electrical impulses to adjust the chip, which allowed it to change its state to function like parts of the human brain.
Researchers at Purdue University in the US say they have successfully created a chip that can be reprogrammed on demand, which could lead to the creation of a computer that can learn more continuously like the human brain.
In a study published in the journal Science, the researchers looked to create a computer chip that could adapt and learn in a way that mimics our brains by using a material called perovskite nickelate, which is very sensitive to hydrogen.
By applying electrical impulses at different voltages, the team was able to adjust the concentration of hydrogen ions on the chip within nanoseconds, creating states that could be mapped out to corresponding functions in the brain.
The human brain is able to go through vast amounts of unstructured data, learn from experiences and rapidly process information with extreme efficiency. Part of this is due to neuronal plasticity, or the ability of the brain to change its structure and function as needed.
“If we want to build a computer or a machine that is inspired by the brain, then correspondingly we want to have the ability to continuously program, reprogram and change the chip,” said Prof Shriram Ramanathan of Purdue’s School of Materials Engineering.
The researchers, in collaboration with Santa Clara University and Portland State University, found a way that electronic chips could dynamically rewire themselves, being able to mimic resistors, memory capacitors, neurons and synapses.
“Using our reconfigurable artificial neurons and synapses, simulated dynamic networks outperformed static networks for incremental learning scenarios,” the team said in the study. “The ability to fashion the building blocks of brain-inspired computers on demand opens up new directions in adaptive networks.”
The researchers are now working to demonstrate these concepts on large-scale test chips that could be used to build a brain-inspired computer. Ramanathan believes that the tech could be readily adopted by the semiconductor industry as the device was built using standard fabrication techniques and was able to operate at room temperature.
The team collaborated with scientists at Argonne National Laboratory, the University of Illinois Chicago, Brookhaven National Laboratory and the University of Georgia, who conducted measurements of the device’s properties.
The research was supported by the US Department of Energy Office of Science, the Air Force Office of Scientific Research and the US National Science Foundation.
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.
