Detail of the student project

List
Topic:Ensembles of graph convolutional networks for semi-supervised classification
Department:Katedra kybernetiky
Supervisor:Georgios Tolias, Ph.D.
Announce as:Diplomová práce, Bakalářská práce, Semestrální projekt
Description:Graph-based method for label propagation form the classical approach for semi-supervised classification. A variant of convolutional neural networks which operate directly on graphs, called graph convolutional networks, is a modern and more effective approach to the task. The goal of this project is to improve the classification accuracy by an ensemble of such networks.
Bibliography:Kipf, Welling, SEMI-SUPERVISED CLASSIFICATION WITH GRAPH CONVOLUTIONAL NETWORKS, ICLR 2017
Responsible person: Petr Pošík