Our paper on ``Federated learning of predictive models from federated Electronic Health Records''...
The Paschalidis Network Optimization & Control (NOC) Lab
Research in the NOC Lab deals with fundamental aspects of optimizing the design and operation of networks, as well as, designing control algorithms to regulate their operation. Increasingly, much of our research is data-driven and it includes important problems in data science and learning theory.
Our view of networks is very broad, encompassing all systems – engineered or natural – with interconnected components. Networks are pervasive in a plethora of application domains, from computer, communication, and sensor networks to supply chains, transportation networks, cyber-physical systems, and networks of autonomous agents. Additionally, naturally occurring networks include biological networks (such as, protein interaction and metabolic networks), social networks, and economic networks.
Research topics currently studied in the lab include: learning from data, reinforcement learning, anomaly detection, transportation networks, computational neuroscience, protein docking, metabolic networks, medical informatics, and energy systems.
A significant percentage of our recent work has focused on computational, algorithmic approaches for problems related to biology and medicine. This includes predictive and prescriptive health analytics, natural language processing of medical text, algorithms for autonomous systems inspired by neuroscience, metabolic networks, and protein docking.
If you are interested in joining the lab and are already at Boston University, contact us. If you are interested in applying to Boston University for graduate studies, please visit this informational page.