January 27, 2016
Computational Network Toolkit Now Available Through GitHub
Microsoft is opening one of its toolkits to developers outside its own walls.
The kit, called Computational Network Toolkit, has been made available on GitHub for those developers hoping to make advancements in artificial intelligence within computers. That’s what Microsoft’s team used it for when looking for ways to make computer recognize human speech in a faster capacity.
Xuedong Huang, Microsoft’s chief speech scientist, said the kit was necessary to create because the tools he and his team had available at the time were slowing them down.
“The CNTK toolkit is just insanely more efficient than anything we have ever seen,” Huang stated.
The team’s efforts were outlined in a recent blog post. That post explains how the field of deep learning has exploded as more researchers have started running machine learning algorithms using deep neural networks, which are systems that are inspired by the biological processes of the human brain. Many researchers see deep learning as a very promising approach for making artificial intelligence better.
That has been the approach of Microsoft researchers who have used CNTK on a set of powerful computers that use graphics processing units, or GPUs.
So, with Microsoft’s internal team able to use the toolkit to fix issues it faced, Huang said it only made sense to let other developers to take the kit to create further advances.
That’s why they decided to make the tools available via open source licenses to other researchers and developers.
W. Brice McVicar is a staff writer for SiteProNews.