The printing of circuits on nanomagnets is a novel way to create computing.
A new research project that creates a rare material known as spin-glass artificially could lead to a paradigm shift in artificial intelligence, allowing for algorithms to be printed directly as hardware. Spin glass’s unusual properties enable AI to recognize objects based on partial images, much like the brain. It also shows promise for low power computing and other fascinating capabilities.
Our work achieved the first experimental realisation of an artificial spin-glass consisting of nanomagnets that are arranged in a way to mimic a neural net,\” said Michael Saccone. He is a postdoctoral researcher at Los Alamos National Laboratory, and the lead author of a new paper published in Nature Physics. Our paper provides the foundation we need to apply these physical systems in practice.
Spin glasses allow you to visualize material structure in a mathematical way. Saccone explained that the ability to manipulate the interaction of these systems by using electron beam lithography for the first-time allows a wide range of computing problems to be represented in spin-glass network.