BeQuiet Jubiläum Banner 970x90
Breaking News

Razer unveils the Viper Mini Signature Edition – a magnesium alloy gaming masterpiece CORSAIR Launches New VENGEANCE a8100 and i8100 Gaming PCs and New Component Products for PC Builders Sonos and Amazon to expand access to voice control with Alexa on Sonos devices Samsung announces new S23/Plus/Ultra series and New Galaxy Book3 Ultra PS5 Beta Introduces Discord Voice Chat, VRR support for 1440p, and more

logo

  • Share Us
    • Facebook
    • Twitter
  • Home
  • Home
  • News
  • Reviews
  • Essays
  • Forum
  • Legacy
  • About
    • Submit News

    • Contact Us
    • Privacy

    • Promotion
    • Advertise

    • RSS Feed
    • Site Map

Search form

Sony Achieves Record Deep Learning Speeds Through Distributed Learning

Sony Achieves Record Deep Learning Speeds Through Distributed Learning

Enterprise & IT Nov 13,2018 0

Sony said that by utilizing its deep learning development framework "Core Library: Neural Network Libraries" in addition to the AI Bridging Cloud Infrastructure (ABCI), it has achieved the fastest deep learning speeds in the world.

Deep learning is a method of machine learning which uses neural networks modeled after the human brain. By harnessing deep learning, image and sound recognition capabilities have seen rapid growth in recent years, even outperforming humans in certain domains. However, the size of data used in this learning and model parameters used to improve recognition accuracy have been increasing, causing a subsequent rise in calculation times. In some cases, it has taken weeks or even months to conduct a single learning session. Because AI development requires a continuous process of trial-and-error, shortening this learning time is of the utmost importance. To this end, distributed learning using multiple GPUs as a means of shortening learning times is emerging as a popular solution.

When increasing the number of GPUs for distributed learning, there are cases where an increase to batch sizes (the amount of data to be processed at one time) halts the learning process, and other cases where the learning speed actually decreases due to the processing delays caused by data transmission times between GPUs. By utilizing technology that can determine the optimal batch sizes and appropriate number of GPUs based on the current state of the learning process, Sony made it possible to carry out learning even in large-scale GPU environments such as ABCI, and increased transmission speeds between GPUs through data synchronization technology optimized for ABCI's system structure. These technologies were implemented into the "Neural Network Libraries," and used ABCI computing resources provided by AIST's "ABCI Grand Challenge" to carry out learning. As a result, it was able to complete ImageNet/ResNet-50 (the general industry benchmark used to measure distributed learning speeds for deep learning) in approximately 3.7 minutes (when using as many as 2,176 GPUs), achieving the world's fastest speeds to date.

The results of this research have been published at "ImageNet/ResNet-50 Training in 224 Seconds (PDF)."

Moving forward, Sony will continue development on related technologies and seek to contribute to the development of society using AI technology.

Tags: deep learningSony
Previous Post
SoftBank to Raise $21 Billion in IPO to Fund Tech Deals
Next Post
China Erases Thousands of Social Media Accounts

Related Posts

  • Sony unveils 2023 line-up of AV receivers with HDMI 2.1

  • PlayStation Plus Monthly Games for February: Evil Dead: The Game, OlliOlliWorld, Destiny 2: Beyond Light, Mafia: Definitive Edition

  • DualSense Edge wireless controller launches globally today

  • PlayStation VR2: 13 new titles and launch lineup revealed

  • Sony Announces Development of New 300mm F2.8 and new Ultra-Wide FE 20-70mm F4 G Lenses

  • The Last of Us premieres today on HBO

  • Introducing Project Leonardo for PlayStation 5, a highly customizable accessibility controller kit

  • PlayStation Plus Monthly Games for January: Star Wars Jedi: Fallen Order, Fallout 76, Axiom Verge 2

BeQuiet Jubiläum Banner 300x600

 

Latest News

Razer unveils the Viper Mini Signature Edition – a magnesium alloy gaming masterpiece
PC components

Razer unveils the Viper Mini Signature Edition – a magnesium alloy gaming masterpiece

CORSAIR Launches New VENGEANCE a8100 and i8100 Gaming PCs and New Component Products for PC Builders
PC components

CORSAIR Launches New VENGEANCE a8100 and i8100 Gaming PCs and New Component Products for PC Builders

Sonos and Amazon to expand access to voice control with Alexa on Sonos devices
Consumer Electronics

Sonos and Amazon to expand access to voice control with Alexa on Sonos devices

Samsung announces new S23/Plus/Ultra series and New Galaxy Book3 Ultra
Smartphones

Samsung announces new S23/Plus/Ultra series and New Galaxy Book3 Ultra

PS5 Beta Introduces Discord Voice Chat, VRR support for 1440p, and more
Gaming

PS5 Beta Introduces Discord Voice Chat, VRR support for 1440p, and more

Popular Reviews

Withings Thermo Wi-Fi-connected temporal thermometer

Withings Thermo Wi-Fi-connected temporal thermometer

EnGenius ECW230 Access Point

EnGenius ECW230 Access Point

Withings Body Plus Scale

Withings Body Plus Scale

Withings Sleep Analyzer

Withings Sleep Analyzer

Noctua NH-D12L CPU Cooler

Noctua NH-D12L CPU Cooler

EnGenius ECW230S AP

EnGenius ECW230S AP

Scythe Fuma 2 CPU Cooler

Scythe Fuma 2 CPU Cooler

be quiet! Pure Rock 2 FX

be quiet! Pure Rock 2 FX

Main menu

  • Home
  • News
  • Reviews
  • Essays
  • Forum
  • Legacy
  • About
    • Submit News

    • Contact Us
    • Privacy

    • Promotion
    • Advertise

    • RSS Feed
    • Site Map
  • About
  • Privacy
  • Contact Us
  • Promotional Opportunities @ CdrInfo.com
  • Advertise on out site
  • Submit your News to our site
  • RSS Feed