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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this work, we quantize a trained Transformer machine language translationmodel leveraging INT8/VNNI instructions in the latest Intel$^\circledR$Xeon$^\circledR$ Cascade Lake processors to improve inference performance whilemaintaining less than 0.5$\%$ drop in accuracy. To the best of our knowledge,this is the first attempt in the industry to quantize the Transformer model.This has high impact as it clearly demonstrates the various complexities ofquantizing the language translation model. We present novel quantizationtechniques directly in TensorFlow to opportunistically replace 32-bit floatingpoint (FP32) computations with 8-bit integers (INT8) and transform the FP32computational graph. We also present a bin-packing parallel batching techniqueto maximize CPU utilization. Overall, our optimizations with INT8/VNNI deliver1.5X improvement over the best FP32 performance. Furthermore, it reveals theopportunities and challenges to boost performance of quantized deep learninginference and establishes best practices to run inference with high efficiencyon Intel CPUs.

Efficient 8-Bit Quantization of Transformer Neural Machine Language Translation Model