Welcome to the Optalysys newsletter – your regular update on the latest developments in optical computing.
In March we announced that we successfully implemented the world’s first convolutional neural network (CNN) using optical processing. This is a rapidly expanding area of Machine Learning. CNNs are used to perform image recognition and analysis tasks such as those in pivotal applications like autonomous vehicles, security and medical image analysis.
These models are computationally extremely intensive, particularly for complex models, where there can be many convolutional “layers” to process. See press release and interview with CEO Dr Nick New in the New Scientist.
Since this announcement we have continued developing the Deep Learning application so that it can now support multiple layer convolutional neural networks. We also have hired our first Deep Learning architect who joins us in October and are looking for further recruits to build out the team for this hugely exciting application.
In June we presented our technology at the Artificial Intelligence conference COGX in London. CogX was a defined as “all things AI” and had over 7000 attendees and 300 speakers. We received a very high level of excitement and interest from end-users, investors and potential employees.
Optalysys in the Press
Beta Programme Cloud System
The Beta test program officially began in February and shows significant potential in the area of genomic assembly. Tests are anticipated to conclude this month and findings published in Q4 this year.
Scientific Advisory Board Inaugural Meeting
The Scientific Advisory Board (SAB) held its inaugural meeting on 23rd July, bringing together Professor Wilkinson (University of Cambridge), Professor Kell (University of Manchester) and Dr Juday (retired NASA scientist) for the first time. The meeting discussed the strategic direction of the technical developments and roadmap.
ESCAPE Project Update
The ESCAPE project is coming to a successful close, with collation of the findings in preparation for the final meeting in November. Over the course of the project, we have developed optical implementations of the two spectral transform dwarfs found in numerical weather prediction, as well as delivering practical demonstrations. Furthermore, we have participated in benchmarking the optical approach against traditional approaches, showing the potential improvements in power consumption of the optical method.
Our team continues to grow welcoming four new team members since the last update:
We welcomed Dr Peter Bloomfield in June who joins us as a Software Engineer based in the Yorkshire office. He holds a PhD in Computing and has over 10 years experience in software programming. Past projects included virtual reality medical visualisation and training systems, healthcare data processing, e-learning platform integration, and computer games.
Dr Imran Sheikh joined us in June as an FPGA contractor for the electronics development. He holds a PhD in Embedded Systems from University of Leicester and previously worked in R&D at GE Power as a technical lead for FPGA/SoC design.
Olivia Hammond joined the Yorkshire office in July as an Optics Designer. She holds a degree in Physics from Oxford University.
Lee Odiam will join in October on completion of his PhD as our first Deep Learning Architect.