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

The popularity of neural networks (NNs) spans academia, industry, and popularculture. In particular, convolutional neural networks (CNNs) have been appliedto many image based machine learning tasks and have yielded strong results. Theavailability of hardware/software systems for efficient training and deploymentof large and/or deep CNN models has been, and continues to be, an importantconsideration for the field. Early systems for NN computation focused onleveraging existing dense linear algebra techniques and libraries. Currentapproaches use low-level machine specific programming and/or closed-source,purpose-built vendor libraries. In this work, we present an open source systemthat, compared to existing approaches, achieves competitive computational speedwhile achieving higher portability. We achieve this by targeting thevendor-neutral OpenCL platform using a code-generation approach. We argue thatour approach allows for both: (1) the rapid development of new computationalkernels for existing hardware targets, and (2) the rapid tuning of existingcomputational kernels for new hardware targets. Results are presented for acase study of targeting the Qualcomm Snapdragon 820 mobile computing platformfor CNN deployment.

Boda-RTC: Productive Generation of Portable, Efficient Code for Convolutional Neural Networks on Mobile Computing Platforms