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

With the growing adoption of deep learning for on-device TinyML applications,there has been an ever-increasing demand for efficient neural network backbonesoptimized for the edge. Recently, the introduction of attention condensernetworks have resulted in low-footprint, highly-efficient, self-attentionneural networks that strike a strong balance between accuracy and speed. Inthis study, we introduce a faster attention condenser design calleddouble-condensing attention condensers that allow for highly condensed featureembeddings. We further employ a machine-driven design exploration strategy thatimposes design constraints based on best practices for greater efficiency androbustness to produce the macro-micro architecture constructs of the backbone.The resulting backbone (which we name AttendNeXt) achieves significantly higherinference throughput on an embedded ARM processor when compared to severalother state-of-the-art efficient backbones (>10x faster than FB-Net C at higheraccuracy and speed and >10x faster than MobileOne-S1 at smaller size) whilehaving a small model size (>1.37x smaller than MobileNetv3-L at higher accuracyand speed) and strong accuracy (1.1% higher top-1 accuracy than MobileViT XS onImageNet at higher speed). These promising results demonstrate that exploringdifferent efficient architecture designs and self-attention mechanisms can leadto interesting new building blocks for TinyML applications.

Faster Attention Is What You Need: A Fast Self-Attention Neural Network Backbone Architecture for the Edge via Double-Condensing Attention Condensers