Making Open Work
May 8–9, 2017: Training & Tutorials
May 10–11, 2017: Conference
Austin, TX

Open source AI at AWS and Apache MXNet

Adrian Cockcroft (Amazon Web Services)
11:00am11:40am Thursday, May 11, 2017
Location: Meeting Room 14 (Sponsored)
Average rating: ****.
(4.33, 3 ratings)

What you'll learn

  • Learn why AWS decided to concentrate investment in MXNet


A wide variety of open source frameworks and tools support artificial intelligence and deep learning. Adrian Cockcroft explains how AWS has packaged a number of them—including deep learning frameworks such as Caffe, CNTK, Keras, MXNet, TensorFlow, Theano, and Torch and supporting tools like Jupyter and Anaconda—into an Amazon Machine Image with optimized GPU support. (Amazon committers also contribute to key technologies, including Linux, Xen, and many of the Apache Hadoop related projects.)

AWS decided to concentrate investment in MXNet and working with the other contributors, sponsored the project into the Apache incubator process. Adrian discusses why AWS picked Apache MXNet, explores the main features of the project, and provides a project status update.

This session is sponsored by AWS.

Photo of Adrian Cockcroft

Adrian Cockcroft

Amazon Web Services

Adrian Cockcroft is vice president of cloud architecture strategy at Amazon Web Services, where he focuses on the needs of cloud native and all-in customers and leads the AWS open source community development program. Adrian has had a long career working at the leading edge of technology and is fascinated by what happens next. Previously he was a developer in the UK; worked at Sun Microsystems; was a founding member of eBay Research Labs; directed a team working on personalizing algorithms, served as a cloud architect, helped teams scale and migrate to AWS, and led the open source program at Netflix; and promoted new ideas around DevOps, microservices, the cloud, and containers at Battery Ventures. He’s also written four books, including Sun Performance and Tuning from Prentice Hall. Adrian holds a degree in applied physics from City, University of London.