Neural networks on photonic chips: harnessing light for ultra-fast and low-power artificial intelligence

A study by the Politecnico di Milano in Science

Milan Building extremely efficient neural networks using photonic chips that process light signals is possible. This was proven by a study by the Politecnico di Milano, conducted together with Stanford University and published in the prestigious journal Science.

Neural networks are distributed computing structures inspired by the structure of a biological brain and aim to achieve cognitive performance comparable to that of humans but in a much shorter time. These technologies now form the basis of machine learning and artificial intelligence systems that can perceive the environment and adapt their own behaviour by analysing the effects of previous actions and working autonomously. They are used in many areas of application, such as speech and image recognition and synthesis, autonomous driving and augmented reality systems, bioinformatics, genetic and molecular sequencing, and high-performance computing technologies.

Compared to conventional computing approaches, in order to perform complex functions, neural networks need to be initially "trained" with a large amount of known information that the network then uses to adapt by learning from experience. Training is an extremely energy-intensive process and as computing power increases, the neural networks' consumption grows very rapidly, doubling every six months or so.

Photonic circuits are a very promising technology for neural networks because they make it possible to build energy-efficient computing units. For years, the Politecnico di Milano has been working on developing programmable photonic processors integrated on silicon microchips only a few mm2 in size for use in the field of data transmission and processing, and now these devices are being used to build photonic neural networks.

“An artificial neuron, like a biological neuron, must perform very simple mathematical operations, such as addition and multiplication, but in a neural network consisting of many densely interconnected neurons, the energy cost of these operations grows exponentially and quickly becomes prohibitive. Our chip incorporates a photonic accelerator that allows calculations to be carried out very quickly and efficiently, using a programmable grid of silicon interferometers. The calculation time is equal to the transit time of light in a chip a few millimetres in size, so we are talking about less than a billionth of a second (0.1 nanoseconds)”, says Francesco Morichetti, Head of the Photonic Devices Lab of the Politecnico di Milano.

“I vantaggi delle reti neurali fotoniche sono noti da tempo, ma uno dei tasselli mancanti per sfruttarne pienamente le potenzialità era l’addestramento della rete. È come avere un potente calcolatore, ma non sapere come usarlo. In questo studio siamo riusciti a realizzare strategie di addestramento dei neuroni fotonici analoghe a quelle utilizzate per le reti neurali convenzionali. Il “cervello” fotonico apprende velocemente e accuratamente e può raggiungere precisioni confrontabili a quelle di una rete neurale convenzionale, ma con un notevole risparmio energetico e maggiore velocità. Tutti elementi abilitanti le applicazioni di intelligenza artificiale e quantistiche.” Aggiunge

In addition to applications in the field of neural networks, this device can be used as a computing unit for multiple applications where high computational efficiency is required, e.g. for graphics accelerators, mathematical coprocessors, data mining, cryptography and quantum computers. The Politecnico di Milano is working on this research with the Photonic Devices Lab and with Polifab, the university's micro and nanotechnology centre.

Lo Studio: