Deep Learning Tech Breaks Speed Limits

Fujitsu Laboratories Ltd. has developed a technology to improve the speed of deep learning software. According to the company, it has now achieved the world's highest speed when the time required for machine learning was measured using the AI Bridging Cloud Infrastructure (ABCI) system, deployed by Fujitsu Limited for the National Institute of Advanced Industrial Science and Technology (AIST).

 

With the spread of deep learning in recent years, there has been a demand for algorithms that can execute machine learning processing at high speeds, and the speed of deep learning has accelerated by 30 times in the past two years. ResNet-50(1), a deep neural network for image recognition, is generally used as a benchmark to measure deep learning processing speed, comparing training times using image data from the ImageNet Large Scale Visual Recognition Challenge 2012 (ILSVRC2012), a contest of image recognition accuracy.

Free Monthly Newsletter

Compelling read? Subscribe to FierceEmbeddedTech!

The embedded tech sector runs the market’s trends. FierceEmbeddedTech subscribers rely on our suite of newsletters as their must-read source for the latest news, developments and analysis impacting their world. Sign up today to get news and updates delivered to your inbox and read on the go.

 

Fujitsu Laboratories says it has developed a technology to expand computation volume per GPU without compromising training accuracy. Highly-efficient distributed parallel processing can be provided by appropriately adjusting the learning rate in accordance to the degree of the deep learning training progress. Fujitsu Laboratories confirmed that it beats the previous speed record by more than 30 seconds, completing the training in 74.7 seconds, the world's highest speed. For deeper learning, visit Fujitsu Laboratories.

Suggested Articles

Renesas has introduced an evaluation and prototyping kit that will speed time-to-market to create secure end-to-end IoT cloud solutions for sensor-based…

According to market research firm ABI Research, global shipments of secure embedded hardware will double until 2023, surpassing 4 billion units.

Free Topology 6050 SoC allows two popular industrial protocols to communicate simultaneously