Gemma Moran

Gemma Moran is an Assistant Professor in the Statistics Department at Rutgers University. Her office is located at Hill Center 475, 110 Frelinghuysen Road, Piscataway, NJ 08854. She teaches courses including MSDS597 Data Wrangling with Python, MSDS534 Statistical Learning for Data Science, and STAT588 Data Mining.

Dr. Moran received her PhD in Statistics from the University of Pennsylvania, where she was advised by Edward George and Veronika Rockova. Prior to joining Rutgers, she was a postdoctoral researcher at the Columbia Data Science Institute working with David Blei.

Dr. Moran's research focuses on developing flexible Bayesian models for analyzing high-dimensional data. Her recent work includes developing identifiable and interpretable deep generative models, particularly variational autoencoders, and creating improved tools for Bayesian model criticism. Her methodological contributions bridge statistics, machine learning, and data science, with applications spanning computer vision, biomedical imaging, and materials science.

Analysis of Dr. Moran's publication record reveals a strong emphasis on Bayesian methodology, particularly in variable selection (spike-and-slab approaches), deep generative modeling, and model criticism. Her work consistently addresses the challenge of maintaining statistical rigor while developing practical machine learning solutions, with a growing focus on model interpretability and identifiability in complex models.

Dr. Moran actively mentors students at multiple levels. Prospective PhD students are encouraged to apply to the Rutgers Statistics PhD program and mention her name. For Masters of Data Science students, she supervises capstone projects requiring independent revisions of group work using different statistical methods or datasets. She also encourages students interested in research to take her MSDS-534 Statistical Learning for Data Science course as preparation for research projects.