Research
Data Science faculty are grouped into clusters of research in four core thematic areas.
Artificial Intelligence
Research in Artificial Intelligence at our institution covers a broad spectrum of foundational and applied topics, including Causal Inference, Uncertainty Quantification, Simulation, Information Interdependence, Graph Models, Streaming Algorithms, and Optimization. These areas are central to developing robust and interpretable machine learning models that can be applied across various domains. Our faculty collaborate closely to push the boundaries of AI, contributing to cutting-edge advancements that have real-world impact, from improving healthcare outcomes to enhancing the reliability of nuclear and particle physics experiments.
Theoretical and Applied Experiments Leveraging LLMs, Generative AI & Reinforcement Learning
C. Fanelli, H. Chen, T. Ford , J. Wang
Geometric and Geospatial Machine Learning Techniques: Application & Theory
Y. He, D. Runfola, A. Nwala, T. Ford, A. Stefanidis
Developing New Techniques for Transfer Learning, Unfolding and Domain Adaptation
J. Wang, D. Runfola, C. Fanelli, T. Ford
Theoretical and Synthetic Approaches to Data Generation
H. Chen, Y. Xiong, C. Fanelli
Data & Society
There are a number of active research groups exploring the use of data to model and understand human activities. This includes applied and simulated programs in social network analysis, human-computer and human-information interactions, the computational aspects of digital humanities, computational social science aspects, and ethical issues associated with the proliferation of AI. Our researchers are focusing on leveraging large datasets to examine patterns in communication, behavior, and cultural trends, as well as developing frameworks to address the societal impacts of data-driven technologies. The integration of data science with social sciences provides critical insights into the dynamics of modern societies, including issues related to privacy, bias, public engagement with information, and the equitable distribution of technology.
Socioeconomic Pattern Prediction Experiments and Applications
D. Runfola, J. Swenson, H. Chen, A. Nwala, A. Stefanidis
Theoretical and Simulated Models of Social Behavior
T. Ford, Y. Xiong
Studying Patterns of Human Interactions with Online Information
Nwala, A. Stefanidis, T. Ford
Applications of AI and Machine Learning to improve Healthcare Outcomes
Chen, H., J. Wang, A. Stefanidis
Data & The Physical Sciences
This research emphasizes the intersection of data science and the physical sciences, focusing on applications across domains such as particle physics, condensed matter physics, and high-energy experiments. Researchers leverage machine learning, simulations, and advanced data analytics to uncover patterns in complex physical systems. These efforts span from reconstructing experimental data in high-energy physics to quantifying uncertainties in cutting-edge measurements. By integrating fundamental physical principles with data-driven techniques, these approaches aim to push the frontiers of scientific discovery and enhance the precision of experimental results.
AI for experimental control, design and calibrations
C. Fanelli, P. Moran, K. Suresh
Near real-time deep learning, identification, reconstruction and anomaly detection
C. Fanelli, P. Moran, K. Suresh
Bayesian Inference and Event-Level Uncertainty Quantification
C. Fanelli, K. Suresh
Data & The Environment
This research leverages cutting-edge data science techniques to monitor and analyze environmental changes and their impact on ecosystems and human communities. Example projects include computer vision-powered marine animal tracking, using satellite data to monitor global-scope environmental outcomes, such as deforestation, climate change, and natural disasters, or using satellite data to deduce information on complex patterns of societal activities.
Data-Driven Modeling and Estimation of Migratory Patterns for Resource Planning and Management
Y. He, J. Swenson, Runfola, D.
Satellite-based Monitoring and Observation of Environmental Outcomes
J. Swenson, D. Runfola, H. Chen
Hardware / Machine Learning Interfaces for Environmental Monitoring in Benthic Environments
Y. He, D. Runfola, Y. Xiong
