Our group investigates problems at the intersection of machine learning, sensing systems, and human-centered AI. Our research spans multiple domains with the shared goal of building robust, privacy-friendly, and adaptive intelligent systems.
We develop machine learning and sensing systems for contactless monitoring of human physiology in real-world environments. Our work spans problems such as blood pressure estimation, sleep apnea detection, respiration and snoring analysis, posture recognition, and physiological peak detection, with an emphasis on robustness, interpretability, and deployment in everyday and clinical settings.
We study how intelligent systems can understand human behavior in homes and other everyday environments. Our research includes activity recognition from event-triggered sensors, layout-aware and trajectory-aware modeling, cross-modal alignment, and generative modeling for symbolic sensor streams, with applications in assistive living and smart environments.
We investigate modern machine learning methods, including representation learning, contrastive learning, and foundation-model-based approaches, for complex real-world data. We are interested in building learning algorithms that are efficient, explainable, and transferable across modalities and domains, including scientific and biomedical applications.
We design machine learning methods for decentralized and resource-constrained environments, where data are distributed across users, devices, and sites. Our work addresses robustness, personalization, heterogeneity, privacy, and adaptation.
Our earlier work also includes wireless networking, localization, and underwater sensing systems, including secure localization, communication protocols, and sensing in challenging environments.