Tech giant Microsoft has open sourced its Machine Learning (ML) framework, Infer.NET, for commercial applications. Infer.NET is now available on GitHub under the permissive MIT license for free use in commercial applications. Developers across the world can contribute with their codes on the Infer.NET repository on Github.
Open sourcing an important tool like this represents Microsoft’s evolution from a purely proprietary model to the open source model. To recall, last week Microsoft made MS-DOS fully open source after over 30 years.
Microsoft uses the framework in large number of products including Xbox, Azure and Microsoft Office. Started in 2004 as a research tool, the infer.NET framework is based on a model-model approach to machine learning.
In 2012 Infer.NET even won a Patents for Humanity award for aiding research in epidemiology, genetic causes of disease, deforestation and asthma.
Microsoft said that they have learned a lot about Machine Learning frameworks and solutions since it started working on Infer.NET in 2004. The framework has evolved from a research tool to being the machine learning engine said Microsoft citing an example called TrueSkill 2, which a system that matches players in online video games. Implemented in Infer.NET, it is running live in the bestselling titles Halo 5 and Gears of War 4, processing millions of matches.
Infer.NET is a little different from other open source ML libraries such as Google’s TensorFlow, Apache Spark or Torch. Unlike these ML framework, Infer.NET enables a model-based approach to machine learning. This lets you incorporate domain knowledge into your model. The framework can then build a bespoke machine learning algorithm directly from that model. This means that instead of having to map your problem onto a pre-existing learning algorithm that you’ve been given, Infer.NET actually constructs a learning algorithm for you, based on the model you’ve provided.
Microsoft is looking to engag with the open-source community in developing and growing the Infer.NET framework further. Also, Infer.NET will become a part of ML.NET, the machine learning framework for .NET developers.
In April 2017, Facebook open sourced Caffe2, a lightweight, modular, and scalable deep learning framework developed by Facebook, built on the original Caffe. Caffe2 is a trendy type of artificial intelligence (AI) that has been designed by keeping expression, speed, and modularity in mind.
Later in June of same year, Japanese electronics major SONY had too open-sourced its Neural Network Libraries, a framework meant for developing artificial intelligence solutions with deep learning capabilities.
Earlier in 2016, Chinese Internet search giant Baidu had open sourced its deep learning platform PaddlePaddle. A PaddlePaddle written machine translation program would need less code when compared to other famously used deep learning platforms.
Source – Microsoft