Research Scientist, (None)
Adrian Vladu is a permanent researcher at CNRS, and is currently based in Paris at the Institut de Recherche en Informatique Fondamentale (Université Paris Cité). He works on multiple aspects of convex and non-convex optimization. In his work he combines tools from continuous optimization, convex geometry and linear algebra in order to obtain improved algorithms for classical discrete problems. He is also interested in various aspects of deep learning, such as designing theoretically-grounded training methods that scale well with limited physical resources.