Facebook's Inclusive AI is the process of providing guidelines to help researchers and programmers design datasets, measure product performance, and test new systems through the lens of 'inclusivity’.Tenzin Norzom
Artificial Intelligence (AI), Augmented Reality (AR) and Virtual Reality (VR) were the three broad themes that took over the second day of Facebook’s F8 developer conference.
Speaking about its AI tool and how it helps address challenges across products, the company announced early efforts to a more ‘inclusive AI’ approach to detect content against its policies in a bid to keep the platform safe with lesser supervision.
So, what is inclusive AI?
Facebook stated ‘when AI models are trained by humans on datasets involving people, there is an inherent representational risk’. Owing to this risk, datasets contain limitations, flaws or other issues, and models perform differently for different people.
Hence, Inclusive AI, as Facebook calls it, is ‘the process that provides guidelines to help researchers and programmers design datasets, measure product performance, and test new systems through the lens of inclusivity’.
For vision related AI, this would include dimensions such as skin tone, age, and gender presentation and for voice, it would include dialect, age, and gender.
Inclusivity and Inclusive AI ran as a theme across technology areas of AR and VR.
Facebook’s augmented reality platform, Spark AR engineers is also leveraging Inclusive AI to ensure their software delivers quality AR effects for everyone. Giving examples, Facebook said,
“Some of the effects are triggered by a hand gesture, so the training data includes various skin tones under a variety of lighting conditions to ensure the system would recognize a hand in front of the camera. Oculus engineers are also using this process for voice commands in virtual reality (VR), using representative data across dialects, ages and genders.”
The social platform claims that Inclusive AI in VR is crucial to help people interact, and come together, regardless of physical distance.
So far, the Menlo Park-based social media platform has been working to bring groundbreaking realism to its ‘Avatars’. Christened 'Codec Avatars', Facebook wants people in the future to create lifelike virtual avatars of themselves.
But, now Facebook is also focussing on ‘full-body’ functions of these avatars, it said at the F8 Conference, adding,
“We’re using a layered approach that replicates human anatomy and can automatically adapt to perfectly match any individual’s appearance and unique motion. We design these models from the inside-out, developing a virtual skeleton then layering on the muscular structure, skin and clothing.”
However, the company agreed that it still has a long way to go before this research results in a product.