Researchers from Carlos III de Madrid University in Spain and MIT have devised a neural network architecture that uses a different mix of optics and electronics than previous schemes to accommodate large numbers of neurons.

The architecture leverages the computational strength of electronics and the fast communications abilities of light. It could be useful in systems that require optical input and neural net computation, like those used for robotic vision.

The design makes it possible to scale the system to a very high number of neurons and the system’s optical interconnects allow for fast communications among neurons.

Pan, F. and Zhang, L. New image super-resolution scheme based on residual error restoration by neural networks. Optical Engineering .October 2003. Volume 42, Issue 10, pp. 3038-3046