Our next seminar will be given by Florent Chiaroni, Ph.D. student at L2S. He will talk about generative adversarial networks (GAN) and one possible application for positive unlabeled learning. More specifically, Florent will explain the working principles of GAN’s, and how he applied them in a weakly-supervised learning scenario, where only few labels of a given class are available. This work has been recently accepted for publication at ICIP 2018.

The seminar will be on June 14th, at 14:30, in the room B555 at Telecom Paristech.