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. The tools will allow developers to take advantage of some of Facebook’s deep learning algorithms and use them to create faster, smarter AI.
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.
You may not realize how much Facebook uses artificial intelligence in its day-to-day operation. Every time you see a recommended friend or are given suggestions for who to tag in a photo, you are seeing Facebook’s deep learning AI in action.
“Deep learning AI” is a phrase stuffed with buzzwords, but it is one of the current hot topics in computing. Traditional AI requires programmers to account for hundreds or thousands of different scenarios, but deep learning allows AIs to do most of the work. As the name suggests, deep learning AI learns from past experience and the data it analyzes to streamline the process in the future. The more data the AI takes it, the better it becomes at recognizing patterns.
The company hopes sharing its tools will help optimize progress across the entire deep-learning research landscape.
Images from images.google.com