Peter Orbanz


I am Professor of Machine Learning in the Gatsby Computational Neuroscience Unit at University College London.

I study machine learning problems involving many interacting variables, in particular their symmetries and large-scale behavior. Specific topics are exchangeability and stationarity (which are forms of symmetry), the large-sample theory of ergodic theorems, random graphs and random structures (which may exhibit symmetry), problems involving many particles in physics, and hierarchies of latent variables. These all turn out to be more closely related than it may seem at first glance.

In the past, I was a PhD student of Joachim M. Buhmann at ETH Zurich, a postdoc with Zoubin Ghahramani at the University of Cambridge, and Assistant and Associate Professor in the Department of Statistics at Columbia University. I have spent sabbaticals at the Isaac Newton Institute, in the Theory Group at Microsoft Research New England, and at UC Berkeley.

Research group

Former PhD students and postdocs