Pierre Vandergheynst

Pierre Vandergheynst is a Full Professor at the Swiss Federal Institute of Technology Lausanne (EPFL) in the Department of Electrical Engineering, with a courtesy appointment in Computer and Communication Sciences. He serves as EPFL’s Vice-Provost for Education since 2015 and leads the Signal Processing Laboratory 2 (LTS2). His research spans harmonic analysis, sparse approximations, mathematical data processing, and applications in signal/image processing, computer vision, machine learning, and graph-based data analysis.

• PhD in Mathematical Physics (1998), Université catholique de Louvain
• Postdoctoral Researcher at EPFL (1998-2001)
• Assistant Professor at EPFL (2002-2007)

His research explores geometry/symmetry in high-dimensional data, redundant dictionaries for dimensionality reduction, and computational harmonic analysis on manifolds. Recent work focuses on protein structure modeling, geometric deep learning, and graph-based signal processing.

Key article trends include graph neural networks for protein analysis, geometric deep learning in neuroscience, and structured knowledge priors in neural models. His 2023-2025 publications emphasize interpretable AI, long-range dependencies in graphs, and molecular representation learning.

Scientific Awards:

• IEEE Signal Processing Magazine Best Paper Award (2023)
• Signal Processing Society Best Paper Award (2022)
• Apple ARTS Award (2007)
• De Boelpaepe Prize, Royal Academy of Sciences of Belgium (2009-2010)

He has supervised over 30 PhD theses and contributed to foundational work in graph signal processing, compressive sensing, and geometric deep learning. His lab develops tools for data science on non-Euclidean structures, with applications in medicine, astronomy, and wireless systems.