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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Convolutional neural networks are paramount in image and signal processingincluding the relevant classification and training tasks alike and constitutefor the majority of machine learning compute demand today. With convolutionoperations being computationally intensive, next generation hardwareaccelerators need to offer parallelization and algorithmic-hardwarehomomorphism. Fortunately, diffractive display optics is capable ofmillion-channel parallel data processing at low latency, however, thus far onlyshowed tens of Hertz slow single image and kernel capability, therebysignificantly underdelivering from its performance potential. Here, wedemonstrate an operation-parallelized high-throughput Fourier opticconvolutional neural network accelerator. For the first time simultaneouslyprocessing of multiple kernels in Fourier domain enabled by optical diffractionhas been achieved alongside with already conventional in the field inputparallelism. Additionally, we show an about one hundred times system speed upover existing optical diffraction-based processors and this demonstrationrivals performance of modern electronic solutions. Therefore, this system iscapable of processing large-scale matrices about ten times faster than state ofart electronic systems.

High Throughput Multi-Channel Parallelized Diffraction Convolutional Neural Network Accelerator