Megha
Thukral

Megha is an ML PhD student in the School of Interactive Computing at Georgia Tech. Her research focuses on representation learning for health sensing and on building machine learning models for sensor data and structured clinical records that generalize across users, devices, home layouts, and clinical settings. Her work addresses distribution shift under varying levels of data and label availability, including few-shot transfer for wearable human activity recognition, layout-agnostic smart-home activity recognition, and self-supervised learning and foundation models for physiological signals and electronic health records. She has previously interned at Samsung Research America’s Digital Health Lab and Bloomberg AI Group, and is fortunate to be advised by Prof. Thomas Plötz.

research areas: Representation Learning, Health Sensing, Human Activity Recognition, Wearables, Smart Homes, Physiological Signals, Clinical Data, Self-Supervised Learning, Foundation Models

email: mthukral3@gatech.edu

website: https://meghathukral.github.io/