Counting published papers is a terrible way to measure progress in machine learning, especially at a place like Apple. For more, please login or subscribe
