AWS and Microsoft unveil Gluon: A new library for machine learning

Oct. 13, 2017, 3:51 p.m. By: Kirti Bakshi


On the 12th of October this year, AWS(Amazon Web Services) and Microsoft's collaboration announced Gluon, a new open source deep learning interface which allows the developers to build machine learning models easily and quickly, without compromising anything on performance. Gluon does not confine itself only to the specialists in AI but also has been designed for developers of all abilities and levels that make it stand out right away as it is believed that the potential of machine learning can only be realized if it is accessible to all developers and not limited to just a few.

AWS and Microsoft may have known to be rivals when it comes to competing for business in cloud storage and services, but to break the grounds in areas that are new and where volumes of data make a difference in how efficiently the services work and also in creating systems that are easier to use, they knew that collaboration was the only way. The result of which we have Gluon in front of us doing just that. We can, therefore, say that the tech industries now realize that they will eventually benefit more from collaboration keeping the commercial competitiveness aside. Also, it is definitely not the first time when both AWS and Microsoft have collaborated on initiatives related to AI.

With this interface, the developers can now build machine learning models just by using simple Python API and a range of pre-built, neural network components that make it even easier for developers of all levels to build neural networks using a simple and concise code. It definitely does provide the developers with the best of both worlds as it is not only a programming language that is easy to understand and concise, but there is so much more to it as it also allows the developers to create neural networks on the fly and to change its size and shape dynamically. Among all the advantages that it provides, the distinct ones include Flexible and Imperative Structure, Dynamic Graphs and High Performance.

Gluon thus turns out to be one of the big steps ahead in removing some of the work in developing AI systems by bringing together two of the key components in any deep learning system namely, training algorithms and neural network models. It being an open source project aims at prototyping, building as well as deploying machine learning models for the cloud and mobile apps as well.

The Gluon interface is, therefore, an open AI ecosystem where developers are provided with the freedom of choice. And as we know that Machine learning not only has the ability to transform the way we work, interact and communicate but also has a hand in transforming our lives and to make this happen the right tools are required to be put in the right hands, and the Gluon interface is just a step taken in the same direction.

Github: Gluon