Debaditya Roy

I am a doctoral student from SCS (Software and Computer Division) of KTH and Qamcom Research and Technology (QRTS). My research is on trustworthy machine learning for time-series-based Human Activity Recognition (HAR) problems. Primarily, I focus on uncertainty estimation, confidence calibration, and privacy-preserving mechanisms such as differential privacy and federated learning for time-series-based HAR.