A new deep learning library, released as a joint offering from Microsoft and Amazon Web Services (AWS), makes it easier for developers to begin creating and integrating machine learning models, regardless of skill level.
The Gluon interface, announced in a joint press release on Thursday, lets developers use a Python API and pre-built neural network components to build machine learning models. It will also work with Apache MXNet and Microsoft Cognitive Toolkit (CNTK), the release said.
Neural networks are built on three core components: Training data, a model, and an algorithm, "The algorithm trains the model to understand patterns in the data," the release said. And because of the complexity of each component, the process can take a long time.
SEE: Research: Companies lack skills to implement and support AI and machine learning (Tech Pro Research)
Deep learning engines have popped up in the market to help accelerate the process, but they often require complex code in order to properly create the models they'll use, the release said. Gluon, on the other hand, uses a simple interface and concise code, allowing for simpler training and more experimentation.
"Developers can use the Gluon interface to create neural networks on the fly, and to change their size and shape dynamically," the release said. "In addition, because the Gluon interface brings together the training algorithm and the neural network model, developers can perform model training one step at a time. This means it is much easier to debug, update and reuse neural networks."
Essentially, the two companies built Gluon "so building neural networks and training models can be as easy as building an app," Swami Sivasubramanian, vice president of Amazon AI, said in the release.
According to the release, the companies believe that Gluon can be used to create machine learning models for the cloud, mobile apps, and edge device as well. Speaking of the cloud, it's interesting that AWS and Microsoft—perhaps the top two players in the IaaS market—have partnered on such an endeavor. Eric Boyd, corporate vice president of Microsoft AI and Research, addressed this in the release, saying it's important for the pair to "pool resources to build technology that benefits the broader community."
If successful, Gluon could provide a good entry point for smaller firms to begin experimenting with deep learning and machine learning technologies without a major investment.
Gluon's reference specification is available on Github here.
The 3 big takeaways for TechRepublic readers
- AWS and Microsoft have partnered to release a new deep learning framework called Gluon, that makes it easier for developers of all skill levels to build machine learning models.
- With Gluon, developers use a Python API and pre-built neural network components to build machine learning models, allowing for more experimentation with less complicated code.
- Gluon could be a good entry point for companies and developers to get started with deep learning, integrating machine learning into their other products.
- Explore the future of deep learning and AI in this 4-course bundle (TechRepublic Academy)
- AWS, Microsoft launch deep learning interface Gluon (ZDNet)
- Special report: How to implement AI and machine learning (free PDF) (TechRepublic)
- Nvidia aims to train 100,000 developers in deep learning, AI technologies (ZDNet)
- IBM Research Distributed Deep Learning code breaks accuracy record for image recognition (TechRepublic)
Conner Forrest has nothing to disclose. He doesn't hold investments in the technology companies he covers.
Conner Forrest is a Senior Editor for TechRepublic. He covers enterprise technology and is interested in the convergence of tech and culture.