creators

Michael Bronstein: geometric deep learning at Oxford

Oxford's DeepMind Professor of AI and AI scientific director at Vienna's Aithyra Institute, Michael Bronstein developed early geometric deep learning methods.

Michael Bronstein is a British-Israeli computer scientist known for developing some of the first methods of "geometric deep learning", which generalises neural network architectures to non-Euclidean data such as graphs and manifolds. Born in 1980, he received his PhD from the Technion in 2007 with a thesis on geometric approaches to pattern recognition, and went on to hold the Chair in Machine Learning and Pattern Recognition at Imperial College London (2018-2021) and visiting appointments at Stanford, MIT and Harvard. In 2018 he co-founded Fabula AI, a startup applying graph neural networks to detect the spread of disinformation on social networks; Twitter bought it on 3 June 2019, and he then led the company's Graph Learning Research. Since 2022 he has held Oxford's DeepMind Professorship of Artificial Intelligence. He is the founding Scientific Director for AI at the Aithyra Institute at the Vienna BioCenter, an Austrian Academy of Sciences institute he was appointed to head in September 2024, applying geometric deep learning to biomedicine and protein design, and is an honorary professor at TU Wien and the University of Vienna. He is also chief scientist in residence at Proxima, the proximity-based drug-discovery company formerly called VantAI. He is a Member of Academia Europaea and a Fellow of IEEE, IAPR, ELLIS and the British Computer Society.

Specialization

geometric deep learning, graph neural networks, AI for structural biology, protein design

Country

United Kingdom