A New Molecule Could Enable Ultra-Fast Brain-Like Computers
An international team of scientists discovered a new molecule that could give the fastest decision-making computers an added speed and efficiency boost, a press release from the University of Limerick (UL), one of the participating institutions, explains.
The molecule could allow for the development of a new type of ultra-fast energy-efficient computing architecture, with widespread implications for the internet of things, artificial intelligence, and other fields in the world of computing.
The team of researchers discovered that a simple molecule made from only 77 atoms acts as "a new fundamental electronic circuit element in which complex logic is encoded in nanoscale material properties," the University of Limerick statement explains. The team developed their brain-inspired computing architecture by experimenting with and analyzing the electrical properties of soft crystals grown from the molecules.
The team's findings, published in the journal Nature, detail how Damien Thompson, Professor in Physics at UL made the discovery using computer simulations run through the Irish Centre for High-End Computing supercomputer.
Brain-inspired computing could boost artificial intelligence
The team's simulations showed that the molecule can rapidly and cleanly switch between different states due to the natural asymmetry in its metal-organic bonds, meaning it can perform ultra-fast decision-making. "The new molecular circuitry means the computer-processing unit no longer has to fetch data for every operation it performs, and this saves enormously on time and energy costs," Thompson explained.
"We are excited about the possibilities because the devices show all the hallmarks of brain computing," Thompson continued. "First, a huge number of tiny, identical molecular processors are networked together and work in parallel. More importantly, they show both redundancy and reconfigurability, which means the device can solve problems even if the individual components do not all work perfectly all the time or in the exact same way every time."
Brain-inspired computing has great potential for changing the way we interact with computers, robots, and artificial intelligence. In 2019, for example, a team of neuroscientists from the University of Colorado Boulder unveiled a brain-inspired computer that could read emotions and tell whether an image was happy or sad via machine learning. Professor Thompson explained that his team's new circuit elements could allow for the development of computers that are faster, smaller, and more energy-efficient, crucial developments in our ever increasingly data-driven world.