World’s Fastest Optical Neuromorphic Processor

Xingyuan Xu holds one of the optical microcombs used in achieving the world’s fastest neuromorphic processor for artificial intelligence. (Source: Swinburne U.)

An international team of researchers led by Swinburne University of Technology has demonstrated the world’s fastest and most powerful optical neuro­morphic processor for arti­ficial intelli­gence, which operates faster than 10 trillion operations per second (TeraOPs/s) and is capable of processing ultra-large scale data. This breakthrough represents an enormous leap forward for neural networks and neuromorphic processing in general.

Artificial neural networks, a key form of AI, can learn and perform complex opera­tions with wide applications to computer vision, natural language processing, facial recognition, speech translation, playing strategy games, medical diagnosis and many other areas. Inspired by the bio­logical structure of the brain’s visual cortex system, arti­ficial neural networks extract key features of raw data to predict properties and behaviour with unpre­cedented accuracy and simplicity. Led by Swinburne’s Professor David Moss, Xingyuan (Mike) Xu (Swinburne, Monash University) and Arnan Mitchell from RMIT University, the team achieved an exceptional feat in optical neural networks: drama­tically acce­lerating their computing speed and processing power.

The team demons­trated an optical neuro­morphic processor operating more than 1000 times faster than any previous processor, with the system also processing record-sized ultra-large scale images – enough to achieve full facial image recognition, something that other optical processors have been unable to accomplish. “This breakthrough was achieved with optical micro-combs, as was our world-record internet data speed reported in May 2020,” says Moss, Director of Swinburne’s Optical Sciences Centre.

While state-of-the-art electronic processors such as the Google TPU can operate beyond 100 TeraOPs/s, this is done with tens of thousands of parallel processors. In contrast, the optical system demons­trated by the team uses a single processor and was achieved using a new technique of simul­taneously interleaving the data in time, wavelength and spatial dimensions through an integrated microcomb source. Microcombs are relatively new devices that act like a rainbow made up of hundreds of high-quality infrared lasers on a single chip. They are much faster, smaller, lighter and cheaper than any other optical source.

“In the 10 years since I co-invented them, integrated micro-comb chips have become enormously important and it is truly exciting to see them enabling these huge advances in information communi­cation and processing. Micro-combs offer enormous promise for us to meet the world’s insa­tiable need for information,” Moss says. “This processor can serve as a universal ultrahigh bandwidth front end for any neuro­morphic hardware – optical or electronic based – bringing massive-data machine learning for real-time ultrahigh bandwidth data within reach,” says Xu, Swinburne alum and postdoctoral fellow with the Electrical and Computer Systems Engi­neering Department at Monash University.

“We’re currently getting a sneak-peak of how the processors of the future will look. It’s really showing us how dramatically we can scale the power of our processors through the innovative use of microcombs,” Xu explains. Mitchell adds, “This technology is applicable to all forms of processing and communi­cations – it will have a huge impact. Long term we hope to realise fully integrated systems on a chip, greatly reducing cost and energy consumption”.

“Convolu­tional neural networks have been central to the arti­ficial intelli­gence revolution, but existing silicon technology increasingly presents a bottleneck in processing speed and energy efficiency,” says key supporter of the research team, Damien Hicks from Swinburne and the Walter and Elizabeth Hall Institute. He adds, “This breakthrough shows how a new optical techno­logy makes such networks faster and more efficient and is a profound demons­tration of the benefits of cross-disci­plinary thinking, in having the inspira­tion and courage to take an idea from one field and using it to solve a fundamental problem in another.” (Source: Monash U.)

Reference: X. Xu et al.: 11 TOPS photonic convolutional accelerator for optical neural networks, Nature 589, 44 (2021); DOI: 10.1038/s41586-020-03063-0

Link: Optical Sciences Centre, Swinburne University of Technology, Hawthorn, Australia

Speak Your Mind

*