Infer.NET: a powerful machine learning system is now OpenSource
The machine learning system, which is so steep that it has helped medical professionals to deepen their knowledge of childhood asthma, has now become available to everyone. The source code for the Microsoft-developed Infer.NET system is published on GitHub , and now anyone can use this intelligent machine learning technology in their applications.
This article is on our news portal.
')
This system is unique in that it creates an individual learning algorithm for each specific case, for each specific data model, so developers do not need to try to create programs using widely available standard algorithms that may not be suitable for a particular application. Since the learning algorithm is compiled for a unique data set, the person who created this set can understand why the system works in one way or another.
System Infer.NET helped pick up a pair of players in Halo 5
The Infer.NET system was used to study and predict children's asthma , which led to a “potentially significant clinical outcome”. She also helped pick up a pair of players in Halo 5 and offer filmmakers films that they might like. In addition, Infer.NET is used in Microsoft services, such as Office and Azure.
Publication of the Infer.NET code in open access was the culmination of a long and ambitious project, ”said Jordan Zaikov, lead development engineer from the Cambridge division of Microsoft Research. - Our group from Microsoft Research started developing this system in 2004. Since then, we have learned a lot about scalable and interpretable solutions using machine learning.
Initially, the Infer.NET system was conceived as a research tool, and in 2008 we made it available for academic use. The result was hundreds of published papers describing the use of our system in various fields, from obtaining information to healthcare. In 2012, Infer.NET received the Patents for Humanity Award for research assistance in the fields of epidemiology, genetic causes of disease, asthma, and the disappearance of forests.
Infer.NET developers at Microsoft
Infer.NET can learn when new data arrives, which is useful for systems used by consumers, and can scale: with its help, researchers from Microsoft have been studying information from billions of web pages.
To learn more about Infer.NET, read a Microsoft - published online book , visit the web page with lessons and examples, and download the system itself.
This article is on our news portal.
')
This system is unique in that it creates an individual learning algorithm for each specific case, for each specific data model, so developers do not need to try to create programs using widely available standard algorithms that may not be suitable for a particular application. Since the learning algorithm is compiled for a unique data set, the person who created this set can understand why the system works in one way or another.
System Infer.NET helped pick up a pair of players in Halo 5
The Infer.NET system was used to study and predict children's asthma , which led to a “potentially significant clinical outcome”. She also helped pick up a pair of players in Halo 5 and offer filmmakers films that they might like. In addition, Infer.NET is used in Microsoft services, such as Office and Azure.
Publication of the Infer.NET code in open access was the culmination of a long and ambitious project, ”said Jordan Zaikov, lead development engineer from the Cambridge division of Microsoft Research. - Our group from Microsoft Research started developing this system in 2004. Since then, we have learned a lot about scalable and interpretable solutions using machine learning.
Initially, the Infer.NET system was conceived as a research tool, and in 2008 we made it available for academic use. The result was hundreds of published papers describing the use of our system in various fields, from obtaining information to healthcare. In 2012, Infer.NET received the Patents for Humanity Award for research assistance in the fields of epidemiology, genetic causes of disease, asthma, and the disappearance of forests.
Infer.NET developers at Microsoft
The Infer.NET development team is looking forward to the open-source community for new ideas for further development and expansion of this system.
Infer.NET can learn when new data arrives, which is useful for systems used by consumers, and can scale: with its help, researchers from Microsoft have been studying information from billions of web pages.
To learn more about Infer.NET, read a Microsoft - published online book , visit the web page with lessons and examples, and download the system itself.
Source: https://habr.com/ru/post/425851/
All Articles