Panagiota Birmpa

Panagiota Birmpa is an Assistant Professor in the Department of Actuarial Mathematics & Statistics at Heriot-Watt University's School of Mathematical and Computer Sciences, Edinburgh. Her research bridges advanced mathematical theory with cutting-edge machine learning applications, focusing on uncertainty-aware methodologies for complex data systems. She actively supervises PhD students and maintains strong collaborative ties with international research groups in applied mathematics and computational science.

Her academic credentials include:

• BSc+MSc (integrated master) in Applied Mathematics and Physical Sciences (majors in Analysis and Statistics) from National Technical University of Athens (NTUA), 2011
• MSc in Pure Mathematics from National and Kapodistrian University of Athens (NKUA), 2014
• PhD in Mathematics from University of Sussex, UK, 2018 (Thesis: Quantification of Mesoscopic and Macroscopic Fluctuations in Interacting Particle Systems)

Dr. Birmpa's research program integrates theoretical mathematics with modern computational challenges through seven core domains: Generative modeling, Scientific Machine learning, Uncertainty Quantification, Probabilistic Graphical models, Interacting Particle Systems, Optimal transport Theory, and Partial Differential Equations. Her interdisciplinary approach enables innovative solutions for complex data analysis problems across scientific domains, particularly where traditional statistical methods face limitations in high-dimensional spaces.

Analysis of her publication trajectory reveals a progression from foundational statistical physics (2017-2018 interface dynamics research) toward contemporary machine learning applications (2021-2024). Her recent work demonstrates increasing sophistication in merging deep learning architectures with uncertainty quantification frameworks, especially for scarce high-dimensional data scenarios where conventional approaches fail. This evolution reflects broader trends in mathematical data science toward robust, interpretable AI systems.

No scientific awards or fellowships are currently listed in her professional profile.

Dr. Birmpa accepts PhD candidates for projects exploring deep learning-graphical model interfaces with uncertainty quantification, building on her prior AFOSR-funded postdoctoral research at UMass Amherst (2021-2022). Her grant history includes significant support from the Air Force Office of Scientific Research for developing particle-based generative algorithms. She maintains active supervision of graduate researchers while pursuing methodological innovations in probabilistic modeling.

Her collaborative research network spans multiple institutions including University of Massachusetts Amherst, with interdisciplinary teams developing novel mathematical frameworks for scientific machine learning. Current projects focus on Lipschitz-regularized gradient flows and generative particle algorithms for high-dimensional data, extending her earlier work on non-equilibrium fluctuations in particle systems.