I am an assistant professor in the Department of Systems and Information Engineering at the University of Virginia.

My research interests lie in data analytics for smart & connected systems, where multiple entities/units (e.g., vehicles, wearable devices, etc.) collect data and are connected through Internet of Things (IoT) technologies. My research aims at the design and application of data-driven methods that transform a connected system into an intelligent system. To achieve this goal, I explore collaborative data analytics based on methods like federated learning, multi-task learning, and Bayesian probabilistic modeling, allowing entities/units to collaborate to establish enhanced smart analytics based on their connectivity. I am dedicated to addressing crucial challenges in collaborative data analytics to enable harnessing heterogeneous, real-time, and dynamic data for smart & connected systems. The methods developed have contributed to diverse smart & connected system applications, including smart manufacturing and digital health.

• Methodologies: Federated analytics, Meta-learning, Multi-task learning, and Bayesian probabilistic modeling
• Applications: IoT-enabled systems in healthcare and manufacturing domains

News

• I like sharing research ideas on data analytics and connected system applications and am happy to discuss collaborative projects. If you find my work interesting or have any interest in potential collaborations, please don’t hesitate to email me!
• Aug 2024: [Grant] I’ll Co-PI the project “ReDDDoT Phase 1: Planning Grant: Facilitating Responsible, Ethical, and Explainable Ergonomic Exposure Assessments when using Artificial Intelligence Methods” sponsored by National Science Foundation (NSF).
• Jul 2024: [Grant] I’ll Co-PI the project “Contributing to Responsible Artificial Intelligence (AI)-Based Biomechanical Exposure Assessment” sponsored by National Safety Council (NSC).
• Jul 2024: [New paper] Check out our paper “Federated Automatic Latent Variable Selection in Multi-output Gaussian Processes”, joint work with my student Jingyi Gao. This paper explores an approach that automatically selects only needed latent variables to extract common latent patterns across outputs under federated settings.
• Apr 2024: [Best paper award] Our paper “Real-time Adaptation for Condition Monitoring Signal Prediction using Label-aware Neural Processes” has been nominated as a finalist of the 2024 IISE QCRE Best Paper Competition! Looking forward to the presentation at 2024 IISE Annual Conference!
• Apr 2024: [Grant] I’ll be serving as the PI of the project “Achieving the Future of Worker Injury Risk Assessment: Personalized and Privacy- Preserving” sponsored by 4-VA.
• Mar 2024: [New paper] Check out our new paper “Real-time Adaptation for Condition Monitoring Signal Prediction using Label-aware Neural Processes” - This paper proposes an extrapolation approach for time-series data featuring real-time adaptation of predictions to online data.
• Nov 2023: I gave a talk entitled “Probabilistic predictive analytics for collaborative systems” at AIML Seminar at University of Virginia.
• Oct 2023: I gave a talk entitled “Fast Personalization for Heterogeneous Condition Monitoring Signals using Neural Processes” at 2023 INFORMS Annual Meeting.
• Oct 2023: I gave a talk entitled “Probabilistic predictive analytics for collaborative systems” at Industrial and Systems Engineering at NCAT state University.
• Jul 2023: [New paper] Our paper “Federated Multi-output Gaussian Processes” has been accepted by Technometrics!
• May 2023: [New paper] Our paper “Federated Condition Monitoring Signal Prediction with Improved Generalization” has been accepted by IEEE Transactions on Reliability!