My current research is focused towards the design of novel and more engaging machine learning paradigms for personalized, context-sensitive and multi-modal sensing (from audio, visual and physiological signals) of human behavior. The aim of these models is to improve the personal medicine and healthcare, as well as the existing wellbeing technologies. The project I am working on at the moment and am particularly passionate about aims to build "the engagement component" for humanoid robots (such as NAO) in order to facilitate educational therapies for children with Autism Spectrum Conditions (ASC), and, in general, to achieve a more socially intelligent human-robot interaction.