Boubekki, Ahcene; Kampffmeyer, Michael; Brefeld, Ulf; Jenssen, Robert (Journal article; Tidsskriftartikkel; Peer reviewed, 2021-06-21)
Deep embedded clustering has become a dominating approach to unsupervised categorization of objects with deep neural networks. The optimization of the most popular methods alternates between the training of a deep autoencoder and a k-means clustering of the
autoencoder’s embedding. The diachronic setting, however, prevents the former to beneft
from valuable information acquired by the latter. In ...