Adrian VLADU is a permanent researcher at the Institut de recherche en informatique fondamentale (IRIF), affiliated with the CNRS and the Université de Paris. Prior to joining the IRIF, he was a Postdoctoral Fellow at Boston University.

His research focuses on multiple aspects of convex and non-convex optimization. His work so far combined tools from continuous optimization and convex geometry in order to obtain improved algorithms for classical discrete problems. He is also interested in various aspects of Deep Learning, such as designing improved training methods and understanding how these affect generalization properties.

Adrian VLADU will present a paper, joint with joint with Kyriakos Axiotis, Adam Karczmarz, Anish Mukherjee, and Piotr Sankowski, Decomposable Submodular Function Minimization via Maximum Flow (read abstract, PDF 31.43 KB) at the EQUIPRICE Lunch Seminar series, organised by Alfred GALICHON in the framework and with funding from his ERC-CoG project EQUIPRICE (Grant agreement No.866274).

More about Adrian VLADU and his research
More about the Equilibrium Methods for Resource Allocations and Dynamic Pricing (EQUIPRICE) Project

Date: THURSDAY, May 19th - 12 PM
Location: via Zoom (link to follow)