Integrated photonics

Artificial Intelligence arises from harnessing computational mechanisms that aim to replicate the biological behaviour of the brain by adapting and learning from large amounts of data. The structures that carry out these processes are called Artificial Neural Networks (ANNs) and are composed of a series of layers that include artificial neurons. An ANN can be trained by means of large datasets and, from there, acquire the ability to infer conclusions from a priori unknown input data.

Photonic technologies, despite the outstanding capabilities and potential they have demonstrated for Artificial Intelligence, have high potential for high bandwidth communications.

The main focus will be on the optimisation of basic functionalities, such as the weighting process, and on the proposal of novel architectural schemes that offer a clear competitive advantage in terms of speed, power consumption, capacity and scalability in relation to the state of the art.