Mr Dimitris Spathis
I am a doctoral researcher in Computer Science at the University of Cambridge, where I am advised by Prof. Cecilia Mascolo.
While our online behaviour is constantly analysed, we lack tools that capture and make sense of our offline behaviour at scale. Wearables and smarphones generate massive traces of real-world behaviour and health, but gaining insights requires a combination of new machine learning methods and understanding data limitations.
Unlike recent advances of deep learning in vision (e.g. self-driving cars), speech (e.g. Alexa) and language (e.g. Google Translate), these non-trivial lifestyle datasets require different modelling due to their dimensionality, noise, sparsity, and multi-modality. I work towards addressing these problems with new models that learn useful representations which can understand and predict your activity, mental health, fitness, sleep, and respiratory health. My research is supported by Jesus College Cambridge, the EPSRC, and the ERC.
Previously, during my studies in Computer Science and AI, I had the chance to work at diverse industries including multinational telcos (Telefonica Research), internet startups (Qustodio), retail tech companies (Ocado), and research labs.
Machine Learning / Deep Learning
Unsupervised / Transfer Learning
Health Data Science