My name is Marzieh, I am a mathematician passionate about uncovering the deep mathematical foundations of data science and machine learning. My research explores the rich interplay between geometry, topology, spectral analysis, and stochastic processes, aiming to reveal how these disciplines can sharpen our understanding of data and improve the performance of learning algorithms. In particular, I work on unifying geometric and topological frameworks, spanning discrete and generalized curvature notions, Morse/Floer/Conley/Forman homology theories, Hodge spectral methods, and stochastic process, into a coherent methodology that advances fundamental mathematics while driving impactful applications in data science, deep learning, biology and beyond.
You can contact me via meidi@mis.mpg.de.