Giovanni Poli’s personal website for Academia

About

My research lies at the intersection of Bayesian methodology and biomedical applications, with a focus on modeling heterogeneity in complex data. During my Ph.D., I developed Bayesian nonparametric methods for principled borrowing of information across related datasets, addressing key limitations of standard statistical approaches.

My work spans multiple areas, including multiple testing in high-dimensional biological data, meta-analysis of heterogeneous studies, and the modeling of time-series. More broadly, my research aims to design flexible yet interpretable statistical models, grounded in Bayesian inference, to support biomedical research.

A central aspect of my research is bridging the gap between advanced statistical methodology and its practical use in clinical research. I am particularly interested in developing methods that remain both theoretically rigorous and accessible to non-specialist users, with the goal of improving reproducibility and interpretability in applied settings.

Research Interests

• Bayesian nonparametrics
• Statistical modeling of heterogeneity
• Clinical data analysis and biostatistics
• Reproducibility and interpretability in statistics
• Applied Bayesian methods for biomedical research

Current Work

My current research focuses on developing novel Bayesian frameworks that integrate flexible modeling with practical usability. This includes designing methodologies that can work alongside standard statistical procedures while providing a principled quantification of uncertainty.

The work is motivated by real-world clinical applications, developed in collaboration with clinical researchers, with a particular focus on oncology and neuroscience.

Ceredits

This website is powered by the Academic Pages template and hosted on GitHub pages.

More info about configuring Academic Pages can be found in the guide, the growing wiki, and you can always ask a question on GitHub. The guides for the Minimal Mistakes theme (which this theme was forked from) might also be helpful.