معرفی
Melanie Weber is an Assistant Professor of Applied Mathematics and Computer Science at Harvard University's John A. Paulson School of Engineering and Applied Sciences (SEAS), leading the Geometric Machine Learning Group. Her research focuses on leveraging geometric structures in data for designing efficient machine learning and optimization algorithms with theoretical guarantees. She holds a PhD from Princeton University (2021) and has held fellowships at the Mathematical Institute of Oxford, Brasenose College, and the Simons Institute. Her work bridges geometry, optimization, and machine learning, with funding from NSF, Sloan Foundation, and Harvard initiatives.
Education:
- PhD in Applied Mathematics, Princeton University (2021)
- BSc/MSc in Mathematics and Physics, University of Leipzig (2016)
Research Interests:
Dr. Weber's research integrates geometric principles into machine learning and optimization, focusing on non-Euclidean spaces, graph structures, and manifold-based methods. Key areas include optimization on Riemannian manifolds, curvature-based analysis (e.g., Ricci curvature), and developing algorithms resilient to data geometry challenges like over-smoothing in graph neural networks. Her work emphasizes theoretical foundations while addressing practical scalability in high-dimensional data.
Awards & Recognition:
- 2024 Sloan Research Fellowship
- 2023 Leslie Fox Prize in Numerical Analysis
- 2023 NSF Grant for Geometric Optimization
Grants & Funding:
Supported by National Science Foundation (NSF), Alfred P. Sloan Foundation, Aramont Foundation, Harvard Dean’s Fund, and Harvard Data Science Initiative.
Labs & Collaborations:
Leads the Geometric Machine Learning Group at SEAS, collaborating with institutions like MIT, Max Planck Institute, and industry labs (Facebook, Google, Microsoft). Active in organizing workshops on geometric methods and curvature analysis.
