Michael T. Wojnowicz

prof_pic.jpg

Hi! I’m a research associate at Harvard University in the Department of Biostatistics, advised by Dr. Jeffrey Miller. Previously, I was a postdoctoral researcher with Tufts University Machine Learning, advised by Computer Science Professor Michael C. Hughes, as well as a Data Scientist at Tufts’ Data Intensive Studies Center.

I’m interested in statistical machine learning, particularly scalable Bayesian inference, time series modeling, and methodologies using measure-theoretic probability.

news

May 24, 2024 I presented new work Scalable Bayesian multi-sample changepoint modeling at the Recent Advances in Variational Inference session of the 2024 New England Statistics Symposium.
Sep 5, 2023 I have started a new position as a research associate at Harvard University in the Department of Biostatistics. I will be developing statistical methods and theory for finding patterns in complex biomedical data, under the direction of Dr. Jeffrey Miller.
Jun 20, 2023 Our paper Approximate inference by broadening the support of the likelihood has been accepted for presentation at the 5th Symposium on Advances in Approximate Bayesian Inference. This is joint work with Assistant Professor Michael C. Hughes and Mathematics Ph.D. student Martin Buck.
Apr 25, 2023 I have been awarded a U.S. patent for Detecting malware with deep generative models.
Jan 3, 2023 We have been awarded a U.S. patent for Bayesian continuous user authentication. This is joint work with Mathematician Dinh Nguyen, Ph.D., and Data Scientist Alexander Kohn, Ph.D.
Jul 29, 2022 Invited speaker at Harvard University’s Data to Actionable Knowledge Lab.
Jul 22, 2022 Spotlight talk at ICML 2022: Easy Variational Inference for Categorical Models via an Independent Binary Approximation. Recording is here.

recent publications

  1. AABI
    Approximate inference by broadening the support of the likelihood
    Wojnowicz, Michael, Buck, Martin D,  and Hughes, Michael C
    In Fifth Symposium on Advances in Approximate Bayesian Inference 2023
  2. ICML
    Easy Variational Inference for Categorical Models via an Independent Binary Approximation
    Wojnowicz, Michael T, Aeron, Shuchin,  Miller, Eric L and 1 more author
    In International Conference on Machine Learning 2022
  3. UAI-TPM
    Easy Variational Inference for Categorical Observations via a New View of Diagonal Orthant Probit Models
    Wojnowicz, Michael, Aeron, Shuchin,  Miller, Eric and 1 more author
    In The 4th Workshop on Tractable Probabilistic Modeling 2021