Project Calypso: machine learning for online advertising and recommender systems
Speaker: Charlotte Laclau, post-doctoral researcher at Laboratoire d'Informatique de Grenoble, Grenoble, France
Abstract: Calypso is a FUI project in collaboration with the University of Grenoble Alpes and two companies, Kelkoo and Purch, on recommender systems and online advertising. In this talk, I will present some of the contributions made since the beginning of the project in November 2015, including the release of two large-scale datasets extracted from both companies' online traffic and RecNet, a novel ranking framework for collaborative filtering that jointly learns a new representation of users and items in an embedded space as well as the preference relation of users over pairs of items. Finally, I will present some challenges in taking into account implicit feedback, large-scale optimization and the need for diversity on bias data.