Presented By O'Reilly and Cloudera
Make Data Work
September 26–27, 2016: Training
September 27–29, 2016: Tutorials & Conference
New York, NY

Machine intelligence at Google scale

Kaz Sato (Google)
2:55pm–3:35pm Thursday, 09/29/2016
Data science & advanced analytics
Location: 3D 08 Level: Intermediate
Average rating: *****
(5.00, 4 ratings)

Prerequisite knowledge

  • A basic understanding of neural networks and machine learning
  • What you'll learn

  • Explore Google's large-scale distributed neural network infrastructure that can accelerate model training 10–40x
  • Description

    The largest challenge for deep learning is scalability. With a single GPU server, it takes hours or days to finish training. This doesn’t scale for production service; eventually you’ll need distributed training in the cloud. Google has been working on a large-scale neural network in the cloud for years and has now started sharing the power with developers.

    Kazunori Sato introduces pretrained ML services, such as the Cloud Vision API and the Speech API, and explores how TensorFlow and Cloud Machine Learning can accelerate custom model training 10–40x with Google’s distributed training infrastructure.

    Photo of Kaz Sato

    Kaz Sato


    Kaz Sato is a staff developer advocate on the cloud platform team at Google, where he leads the developer advocacy team for machine learning and data analytics products such as TensorFlow, the Vision API, and BigQuery. Kaz has been leading and supporting developer communities for Google Cloud for over seven years. He’s a frequent speaker at conferences, including Google I/O 2016, Hadoop Summit 2016 San Jose, Strata + Hadoop World 2016, and Google Next 2015 NYC and Tel Aviv, and he has hosted FPGA meetups since 2013.

    Comments on this page are now closed.


    Picture of Kaz Sato
    Kaz Sato
    10/04/2016 12:47pm EDT

    Here you go:

    Picture of André Morrow
    André Morrow
    10/04/2016 12:39pm EDT

    All Strata + Hadoop World 2016 slide presentations have been posted if they were made available to us by the speakers.

    Arun Venkateswaran
    10/04/2016 12:29pm EDT

    Hey Kaz,

    Could you post slides on this session please?