News

AWS, Microsoft Team Up for 'Gluon' Deep Learning Project

• By Gladys Rama
• October 13, 2017

While Microsoft has been doggedly chasing Amazon Web Services (AWS) in the cloud computing arena, the two tech giants have partnered on a new deep learning initiative called Gluon.

It's desribed as an interface that lets developers of varying skill levels use Python and pre-built deep learning templates to simplify the building of models to run neural networks. The two companies yesterday defined Gluon as "an open source deep learning interface [that] allows developers to more easily and quickly build machine learning models without compromising training performance".

Typically, the process of building a neural network can take as long as a week or more, as developers train network models to finely parse large volumes of data. Besides being time-consuming, it's also labor-intensive, often requiring developers to write long lines of complex code that are hard to modify, debug and reuse.

Altogether, the requirements of building a neural network are taxing for even experienced developers and can be prohibitively complicated for beginners. AWS and Microsoft developed Gluon in an effort to remove these barriers.

"At AWS, we've been experimenting with some ideas in MXNet around new, flexible, more approachable ways to define and train neural networks. Microsoft is also a contributor to the open source MXNet project, and were interested in some of these same ideas. Based on this, we got talking, and found we had a similar vision: to use these techniques to reduce the complexity of machine learning, making it accessible to more developers," said Matt Wood, AWS general manager of AI, in a blog post.

According to the two companies, there are four attributes that set Gluon apart from other deep learning tools. First, the Gluon interface lets developers use "simple, clear, concise" code to build networks, according to Wood. Second, it lets them create more "dynamic" networks that are easier to debug and adapt. Third, it enables developers to create "more sophisticated algorithms and models." Fourth, it provides these benefits without slowing down the training process.

Gluon is not the first time that AWS and Microsoft have collaborated around AI; in August, the two companies agreed to let their respective voice-responsive digital assistants, Alexa and Cortana, interact with each other in a bid to improve productivity for their customers. "The collaboration between Microsoft and Amazon reflects our belief that when people and technology work together, everybody wins," Microsoft said of the Cortana-Alexa integration at the time.

The Gluon project appears to have sprung from that same philosophy. "We believe it is important for the industry to work together and pool resources to build technology that benefits the broader community. This is why Microsoft has collaborated with AWS to create the Gluon interface and enable an open AI ecosystem where developers have freedom of choice," said Eric Boyd, corporate vice president of AI and Research at Microsoft, in an announcement.

Available on GitHub here, Gluon currently supports Apache MXNet. A future release will add support for the Microsoft Cognitive Toolkit, as well as other frameworks, according to the two companies.