Semi-Supervised Learning via Compact Latent Space Clustering

Konstantinos Kamnitsas, Daniel C. Castro, Loïc Le Folgoc, Ian Walker, Ryutaro Tanno, Daniel Rueckert, Ben Glocker, Antonio Criminisi, Aditya V. Nori

Abstract: We present a novel cost function for semi-supervised learning of neural networks that encourages compact clustering of the latent space to facilitate separation. The key idea is to dynamically crea...