Tiny Quantum Nanostructures Could Make AI Less of an Energy Hog

University of Wisconsin-Madison researchers designed a theoretical quantum-emitter nanostructure that could supply the nonlinear operations needed for energy-efficient optical neural networks. Simulations indicated the approach could reduce the power needed for optical nonlinearity by seven orders of magnitude compared with conventional optical materials. The work remains computational, but the researchers said advances in diamond-based quantum photonics could make experimental implementation feasible with current technology. PhD student Qingyi Zhou used computational simulations to find a new way to bypass the nonlinearity bottleneck in optical neural networks.…