New ‘super-brain’ learns laws of nature to fast-track optical component design The physics-informed artificial intelligence predicts optical properties in milliseconds. Researchers at Chalmers University of Technology have developed a machine learning system that learns the laws of physics before training, allowing it to design advanced optical materials up to ten times faster than conventional methods. The breakthrough could accelerate the development of optical components used in applications ranging from quantum computing to camera and eyeglass lenses. “When we fed the super-brain information about the laws of physics, it immediately got much smarter. Our calculations now take one tenth of the time previously required,” said Philippe Tassin, professor at the Department of Physics and Astronomy. Designing advanced optical materials The Chalmers team works in nanophotonics, a field focused on controlling and manipulating light at scales smaller than its wavelength. At these dimensions, light behaves differently than it does in conventional optical systems, enabling scientists to create artificial materials with properties not found in nature. Using supercomputer simulations, the researchers design optical materials that could be used to make camera and eyeglass lenses lighter, thinner, and more effective. Their work could also support future developments in quantum technologies. Together with researchers at the Department of Microtechnology and Nanoscience, where Sweden’s first larger quantum computer is being built, the team is exploring whether nanostructured materials can be designed to control how light travels. The concept involves using mechanically compliant photonic crystals to transmit information between quantum computers or across longer distances using optical frequencies. The simulations play a central role in this work, helping researchers determine how materials should be structured to achieve the desired optical properties. Solving the bottleneck The research relies heavily on machine learning and neural networks, which analyze vast amounts of simulation data to predict how materials will behave.