Facebook Trains Artificial Intelligence to "See"

Social media platform Facebook (NASDAQ: FB) has announced recently that it will launch an artificial intelligence program in order to track what users are viewing online. It is a "computer vision" program by the name of SEER that has notably outdone prior artificial intelligence programs.

This particular SEER program also has underwent a "classification accuracy" test and received a score of 84.2%. The test, offered by the company ImageNet, gleans information on whether or not an artificial intelligence program can truly pinpoint the components of a photo.

Facebook's researchers said in a blog post, "The future of AI is in creating systems that can learn directly from whatever information they're given - whether it's text, images, or another type of data - without relying on carefully curated and labeled data sets to teach them how to recognize objects in a photo, interpret a block of text, or perform any of the countless other tasks that we ask it to."

Facebook also stated that SEER ("self-supervised learning") was taught how to identify specific objects in photos through the inspection of various Instagram (which is owned by Facebook) images. Although many Instagram users may be shocked to learn that the images they are posting are being utilized in order to test the artificial intelligence programs for Facebook, Facebook is trying its best to become more communicative and detail-oriented so that other companies can take advantage of the AI program. Some other companies that would like to make usage of this type of AI include Google (NASDAQ: GOOGL) and Microsoft (NASDAQ: MSFT).

In the eyes of Facebook's Chief Scientist Yann LeChun, this opportunity for the social media platform to have artificial intelligence will simply "build machines that have the background knowledge, or 'common sense,'" to take on tasks that go beyond the capabilities of what today's machines have the power to do. In the real world, the chance for artificial intelligence may seem far-fetched, but Facebook has proven such ideas otherwise.

Through the implementation of this new SEER program, Facebook hopes to not only survey what users are looking at, but also to interpret the given information. If Facebook is successful with this program, then it may open the future to new playing fields and visions for tomorrow, in terms of the world of computer research and data at large.