The party is ON! Join us at TNW Conference 2021 in Amsterdam for face-to-face business!

The heart of tech

This article was published on January 16, 2015

Facebook open-sources its deep-learning AI tools

Facebook open-sources its deep-learning AI tools
Napier Lopez
Story by

Napier Lopez

Reporter

Napier Lopez is a writer based in New York City. He's interested in all things tech, science, and photography related, and likes to yo-yo in Napier Lopez is a writer based in New York City. He's interested in all things tech, science, and photography related, and likes to yo-yo in his free time. Follow him on Twitter.

Facebook is sharing some of its technology. The company’s artificial intelligence research team today announced that it is open sourcing its deep-learning AI tools.

The software will be available on the Torch library, which serves as an open-source environment for machine learning development. Torch is widely used for research in academia, as well as by companies like Google, Twitter and Intel.

Facebook claims its deep-learning modules are significantly quicker than the default ones available through Torch, and allow the company to work on larger neural networks in less time.

Notable improvements include a 23.5x speed-up over publicly available convolutional layer codes, and the ability to parallelize neural networks training over GPU cards.

The company hopes sharing its tools will help optimize progress across the entire deep-learning research landscape.

➤ FAIR open sources deep-learning modules for Torch [Facebook Research]