Giorgos Bouritsas

Giorgos Bouritsas is a machine learning scientist serving as a postdoctoral fellow at the Archimedes AI unit and the University of Athens, while also holding an adjunct lecturer position at NCSR Demokritos. He received his PhD in computer science from Imperial College London and his MEng in electrical and computer engineering from the National Technical University of Athens. His academic journey includes research stints at Google DeepMind, École Polytechnique Fédérale de Lausanne, KU Leuven, NCSR Demokritos, and Universitat Politècnica de Catalunya.

Dr. Bouritsas specializes in geometric and graph deep learning, with research focusing on neural network architectures for geometric data, weight space learning, and applications in biology and chemistry. His work bridges theoretical analysis with practical implementations, particularly in developing methodologies for complex networks, physical systems, 3D objects, and neural network weight spaces. He has made significant contributions to graph neural networks, including novel approaches to improving expressivity through subgraph isomorphism counting and developing spiral convolutional networks for 3D shape representation.

His recent publications demonstrate a strong trajectory in geometric deep learning, with papers accepted at top conferences including NeurIPS (with an oral presentation in 2024), ICML, CVPR, and ICCV. His 2024 workshop proposal on Neural Network Weights as a New Data Modality was accepted for ICLR 2025. He regularly serves as a reviewer for major machine learning conferences, earning outstanding reviewer distinctions at NeurIPS, ICML, and LoG.

• Outstanding Reviewer at NeurIPS 2021, 2024
• Outstanding Reviewer at ICML 2022, 2024
• Outstanding Reviewer at LoG 2022, 2023

Dr. Bouritsas teaches Deep Learning in the MSc in AI program at NCSR Demokritos, demonstrating his commitment to academic service and education. His research has practical applications spanning cryo-EM image analysis, 3D facial recognition for medical diagnostics, and theoretical foundations of contrastive learning frameworks.