R21: Knowledge-informed Deep Learning for Apnea Detection with Limited Annotations
(NIBIB Trailblazer Award for New and Early Stage Investigators)
Funding agency: National Institutes of Health Role: PI
Co-PIs: Drs. Feng Liu (Stevens Institute of Technology) and Changyue Song
The goal of this research is to create a domain knowledge-informed framework to enable machine learning with limited data availability to generate comprehensive sleep profile learning and monitoring. The technical objective of the proposed study is to construct weakly supervised deep learning models for real-time sleep event detection based on noisy physiological signals with limited annotations.
RAPID: Adaptive Sampling Strategies for COVID-19 Mass Testing
Funding agency: National Science Foundation Role: PI
Co-PIs: Drs. Jaclyn Hall, Thomas Hladish
Senior Personnel: Dr. Alexander Semenov
The proposal objective is to develop a data-driven strategic framework to optimize COVID-19 mass testing resources and collect community testing data by adaptively sampling within census block groups.
Output: The project generates aggregated COVID-19 datasets about viral and antibody testing in several Florida communities. Detailed descriptions and data can be found at https://www.dropbox.com/sh/xuk6n9529sph6qz/AADcEcjDcyCE6c_uWUapY86ka?dl=0 upon request to the PI.
Data Analytics and System Informatics Enhanced Anomaly Detection and Diagnosis for Manufacturing Infrastructure Cybersecurity
Funding Agency: The Florida Center for Cybersecurity Role: PI
Co-PI: Dr. Devashish Das (University of South Florida)
This project aims to address the interweaved challenges by bridging the knowledge-gaps through to achieve smart cybersecurity-related anomaly detection and fault localization, with a specific focus on applications to manufacturing systems and from the incorporation of data analytics and system-specific domain knowledge.
Safe and Efficient Operations of Autonomous Aircraft for Quick Anomaly Detection
Funding agency: The Florida Space Grant Consortium Role: PI
In this project, the PI proposes data-driven adaptive strategies to safely and quickly identify abrupt and stochastic abnormalities with autonomous aircraft, integrating concepts of machine learning, uncertainty quantification, statistical process control, mathematical and distributed optimization to dynamically inform online decision making.
Socioeconomic Impacts of COVID-19: A Criminological Perspective
Founding agency: UF Informatics Institute Role: co-PI
PI: Dr. Yujie Hu
Accountable Artificial Intelligence: Toward Public Interest-Minded News Recommendation Systems
Funding agency: UF Informatics Institute Role: co-PI
PI: Dr. Jieun Shin
Travel Awards for Research Grant EnhancemenT (TARGET): Factfinding Groundwork for a UF Biofoundry
Funding agency: UF Institute of Food and Agricultural Sciences Dean for Research Office Role: co-PI
PI: Dr. Andrew Hanson