We show that optical waves passing through a nanophotonic medium can perform artificial neural computing. Complex information, such as an image, is encoded in the wave front of in-put light. The medium continuously transforms the wave front to realize highly sophisticated computing tasks such as image recognition. At the output, optical energy is concentrated to well defined locations, which for example can be interpreted as the identity of the object in the image. These computing media can be as small as tens of wavelengths in size and thus offer extremely high computing density. They exploit sub-wavelength linear and nonlinear scatterers to realize sophisticated input-output mapping far beyond traditional nanophotonic devices. To enable these complex neural computing, we draw inspiration from artificial neural network and use stochastic gradient decent to optimize nonlinear nanophotonic structures with structural gradient computed from adjoint state method.
|