A group of researchers from around the world has made a groundbreaking discovery that could transform the artificial intelligence industry.
The team, spearheaded by scientists from the University of Sydney, has successfully demonstrated that nanowire networks have the ability to display both short- and long-term memory, mimicking the functionality of the human brain.
Higher-Order Cognitive Functions
According to Dr. Alon Loeffler, the leader of the team, this research has revealed that non-biological hardware has the potential to emulate higher-order cognitive functions that are typically linked to the human brain.
The team's previous work with nanotechnology resulted in the creation of a brain-inspired electrical device with circuitry similar to neural networks and signaling resembling synapses.
Nanowire networks, composed of minuscule silver wires coated with plastic material, form a type of nanotechnology that resembles the interconnected physical configuration of the human brain.
As a result of this likeness, the potential uses for advancements in nanowire networks are vast and varied, spanning from enhancing robotics to producing sensor devices.
"This nanowire network is like a synthetic neural network because the nanowires act like neurons, and the places where they connect with each other are analogous to synapses," senior author Professor Zdenka Kuncic from the School of Physics said in a statement.
"Instead of implementing some kind of machine learning task, in this study, Dr. Loeffler has actually taken it one step further and tried to demonstrate that nanowire networks exhibit some kind of cognitive function."
Read Also: Google Bard Launches New Feature that Will Help Users Code, For Generation and Debugging
Nanowire Takes a Memory Test
The researchers conducted a memory test on the nanowire network, which was similar to the N-Back task commonly used in human psychology experiments.
The task involved remembering a particular image from a series of images presented in a sequence, and an N-Back score of 7 indicated the ability to recognize the same image that appeared seven steps back.
After conducting the test on the nanowire network, the researchers found that it was able to remember a desired endpoint in an electric circuit seven steps back, achieving a score of 7 in the N-Back test.
Dr. Loeffler explained that they manipulated the voltages of the end electrodes to force the pathways to change and go where they wanted them to go rather than letting the network act on its own.
When they did this, the network's memory had higher accuracy and did not decrease over time, indicating that they had discovered a way to strengthen the pathways and direct them toward the desired endpoint.
The researchers also noted that the nanowire network, like the human brain, consolidates information into memory when it is constantly reinforced. As a result, the need for reinforcement decreases once the information is consolidated into long-term memory.
The study has significant implications for the development of artificial intelligence, as it suggests that it may be possible to create hardware that can learn and remember like the human brain.
The findings of the team were detailed in the journal Science Advances.
Related Article: Researchers Use Machine Learning Model for Accurate Prediction of Strokes
