Current Projects
- Efficient, Private, and Explainable Federated Learning for Financial Crime Detection : We are developing resource-efficient privacy-preserving federated methods for learning and explainiability, with the primary application financial crime detection.
- NSF CAREER: Toward A Machine Learning Framework for the Internet of Things : This goal of this project is to develop a new paradigm and tools for machine learning over the massive-scale, geographically distributed data in the Internet of Things.
- NSF CSR: Small: Virtual Sky: Morphable Geo-Spatial Computing for the Internet of Planes : In this project, we are developing distributed methods for situational awareness and coordination in future aerospace systems.
Past Projects
- Secure and Robust Cross-Silo Vertical Federated Learning : In this project, we developed new Vertical Federated Learning methods that operate over datasets with missing labels and features. This project was funded by the Rensselaer-IBM Artificial Intelligence Research Collaboration (2021 - 2022).
- Towards a General Framework for Robust Vertical Federated Learning : This project focused on Vertical Federated Learning, specifically the creation and analysis of methods that are communication efficient and robust to heterogeneous participants. This project was funded by the Rensselaer-IBM Artificial Intelligence Research Collaboration (2020 - 2021).
- Predictive Resource Configuration and Scheduling for Cost-Efficient Data Processing in AWS (PSEG, 2019)
- Interactive Visualization of Communications in Networked Programs to Enhance Student Learning and Aid Debugging (RPI Teaching and Learning Seed Proposal Award, 2019)
- NSF CSR: III: Small: Collaborative Research: A Hybrid Vehicle-Cloud Solution for Robust, Cost-Efficient Road Monitoring (2015-2019)
- Robust Autonomous Underwater Grasping with the Parallel-Jaw Gripper (2015)
