Global Bigdata Conference

Facebook has a whole set of internal tools to try and optimize its neural networks to run on mobile devices. Still, the company finds it difficult to navigate a smartphone market that is byzantine in its complexity, with thousands of different chipsets, most of poor performance, and software stacks that aren't quite up to the job. AI on mobile devices is a bit of a mess, and it's a headache for Facebook, which gets 90% of its advertising revenue off of people using its service on mobile. Those are some takeaways of a recent research paper from Facebook's AI folks, who detail how they've had to come up with all manner of tricks to get around the hardware shortcomings of mobile. That includes things like tweaking how many "threads" in an application to use to reach a common denominator across a plethora of different chip designs and capabilities. That means they can't generally "optimize" their code for a given device.