English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Convolutional neural networks (CNNs), inspired by biological visual cortexsystems, are a powerful category of artificial neural networks that can extractthe hierarchical features of raw data to greatly reduce the network parametriccomplexity and enhance the predicting accuracy. They are of significantinterest for machine learning tasks such as computer vision, speechrecognition, playing board games and medical diagnosis. Optical neural networksoffer the promise of dramatically accelerating computing speed to overcome theinherent bandwidth bottleneck of electronics. Here, we demonstrate a universaloptical vector convolutional accelerator operating beyond 10 TeraFLOPS(floating point operations per second), generating convolutions of images of250,000 pixels with 8 bit resolution for 10 kernels simultaneously, enough forfacial image recognition. We then use the same hardware to sequentially form adeep optical CNN with ten output neurons, achieving successful recognition offull 10 digits with 900 pixel handwritten digit images with 88% accuracy. Ourresults are based on simultaneously interleaving temporal, wavelength andspatial dimensions enabled by an integrated microcomb source. This approach isscalable and trainable to much more complex networks for demanding applicationssuch as unmanned vehicle and real-time video recognition.

11 TeraFLOPs per second photonic convolutional accelerator for deep learning optical neural networks