Facebook, Microsoft team up for AI interoperability


Facebook and Microsoft joined forces to develop the Open Neural Network Exchange, an open-source standard to improve interoperability between artificial intelligence frameworks.

The ONNX standard brings together multiple AI frameworks, including Cognitive Toolkit, Microsoft's open-source framework for building deep neural networks. Each framework is optimized for different functions, such as network architectures or mobile device inferencing, so developers can "more easily move between frameworks and use the best tool for the task at hand," according to a Microsoft blog post.

The standard aims to overcome a challenge engineers and researchers often face when developing an AI model. Developers might select one AI framework at the start of a project, even if a separate framework proves more viable as the project evolves. With the ONNX standard, however, the developers will be able to transfer their models between frameworks.

"We developed ONNX together with Microsoft to bridge this gap and to empower AI developers to choose the framework that fits the current stage of their project and easily switch between frameworks as the project evolves," according to a Facebook Research blog post. "We are focused on bringing the worlds of AI research and products closer together so that we can innovate and deploy faster."

The companies released the initial version of the ONNX framework Sept. 7 on GitHub.

© Copyright ASC COMMUNICATIONS 2021. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.


Featured Whitepapers

Featured Webinars