Rate My Professor Mike Wojnowicz

Wojnowicz's research lies at the intersection of statistics and machine learning, focusing on scalable Bayesian inference, time series modeling, changepoint detection across multiple samples, and methods that blend interpretable probabilistic models with techniques for handling large, complex datasets. His applications include cybersecurity, cancer research via copy number alterations, stroke recovery prediction, and soldier performance analysis. Key publications include "Discovering group dynamics in coordinated time series via hierarchical recurrent switching-state models" (Transactions in Machine Learning Research, 2025), "Easy Variational Inference for Categorical Models via an Independent Binary Approximation" (International Conference on Machine Learning, 2022, spotlight talk), "Approximate inference by broadening the support of the likelihood" (Symposium on Advances in Approximate Bayesian Inference, 2023), and "Influence sketching: Finding influential samples in large-scale regressions" (IEEE International Conference on Big Data, 2016). At Montana State University, starting January 2025, he contributes to the SMART FIRES project on sensors, machine learning, and AI for real-time fire science, teaches classes on probabilistic data modeling, and seeks students for probabilistic machine learning research. He has presented at the New England Statistics Symposium (2024) and given invited talks, such as at Harvard's Data to Actionable Knowledge Lab (2022).