HomeinetFacebook's new AI algorithm teaches itself to work ...

Facebook's new AI algorithm teaches itself to work with less human help

Most artificial intelligence is still built on a foundation of human endeavor. If you look inside an AI algorithm, you will find something constructed with data edited and labeled by humans.

Now, Facebook has shown how some AI algorithms can learn to do useful work with much less human help. The company created an algorithm that learned to recognize objects in images with a little help from tags.

AI facebook

Facebook's algorithm, called Seer, has been powered by more than a billion images from Instagram, deciding for itself which objects look alike. The algorithm was then given a small number of tagged images. He was then able to identify the images and the algorithm was trained using thousands of highlighted examples of each object.

"The results are impressive," said Olga Russakovsky, an assistant professor at Princeton University who specializes in artificial intelligence (AI) and computer vision. "Acquiring self-supervised learning at work is very difficult and important discoveries in this area have important implications for improved visual recognition. "

Russakovsky says it is remarkable that Instagram images were not manually selected to facilitate independent learning.

Research on Facebook is a milestone in an AI approach known as "self-monitoring learning," says Facebook lead scientist Yann LeCun.

LeCun pioneered the machine learning approach known as deep learning which includes data supply to large artificial neural networks. About a decade ago, deep learning emerged as the best way to program machines to do all sorts of useful things, such as image sorting and speech recognition.

But LeCun says the conventional approach, which requires "training" an algorithm by powering it with multiple signals data, just will not escalated. "I have been supporting this whole idea of ​​self-supervised learning for a long time," he says. "In the long run, advances in artificial intelligence will come from from programs that watch videos all day and learn like children ".

LeCun says that self-directed learning could have many useful applications, for example learning to read medical images without need to mark so many scans and x-rays. He says a similar approach is already being used to automatically create hashtags for images in Instagram. And he says Seer technology could be used on Facebook to match ads to posts or to filter unwanted content.

Facebook research is based on modifying deep learning algorithms to make them more efficient and effective. Self-supervised learning has previously been used to translate text from one language to another, but has been more difficult to apply to images than to words. LeCun says the research team has developed a new way in which algorithms learn to recognize images even when part of the image has modified.

Facebook will release some of the technology behind Seer, but not the algorithm itself, because it was trained using Instagram user data.

Source of information: wired.com

Teo Ehchttps://www.secnews.gr
Be the limited edition.