I’m a PhD student in machine learning at Oxford, supervised by Yee Whye Teh and Arnaud Doucet and funded by Google DeepMind.
My recent research interests include invertible neural networks, flow-based models and continuous depth neural networks. I’m also interested in the intersection of physics and machine learning and especially how we can build models that combine learning from data with physical principles. I’ve also worked on neural rendering during an internship at Apple and am very interested in how we can use graphics renderers to make computer vision algorithms more robust and interpretable.
Previously, I’ve worked on disentanglement, inpainting and its applications to various physical problems.
Before my PhD, I worked for 2 years as a machine learning scientist at STIC in the Bay Area. I studied computational maths at Stanford University and theoretical physics at Imperial College.
My email is emilien.dupont (at) stats.ox.ac.uk