A Bayesian approach towards Coaching A.I.
The Freeletics coaching A.I., is envisioned to be an empathetic, understanding coach that is capable of measuring each user’s capability, motivation and preferences. Achieving this dictates describing the available workouts and exercises in a space that users can also be mapped into. Furthermore, this requires building a probabilistic user model that describes every athlete as a unit and in relation to the complete population. In this talk, a high-level description on how the Coaching A.I. makes use of the multiple interaction points within the Freeletics ecosystems and how it leverages itself to be each athlete's personal coach.
Laith is the Senior Machine Learning Engineer at Freeletics, he is responsible for the research, development and implementation of the coaching A.I. of the Freeletics coach. Working towards his Dr-Ing degree in Cognitive Robotics from the Technical University of Munich, Laith’s research focus has been towards modeling cognitive intention recognition capabilities in robots. He previously worked at the Bristol Robotics Laboratory as a robotics research associate focusing on the field of Intention, plan and action understanding.