Andy’s work sits at the intersection of trustworthy machine learning and health care. He focuses on federated and privacy-preserving learning, in particular what happens to model behaviour when data cannot leave the institution that collected it: heterogeneity across sites, communication constraints, and the gap between reported performance and performance that holds up in a clinical setting.
Before moving to Essen, he spent two years teaching in the Master of Data Science program at the University of British Columbia, covering data visualisation, probability and statistics, databases, and communication. He earned his PhD in neuroscience at the University of British Columbia, working on applied machine learning in addiction psychiatry.
His current projects and topics include FLIP-IT (federated learning for medical data), privacy-preserving aggregation and communication-efficient federated protocols, and the evaluation and reliability of clinical ML models, including how performance transfers across sites.
