An Architecture to Efficiently Learn Co-Similarities from Multi-View Datasets

Gilles Bisson (1) , Clément Grimal (1)

1 AMA - Analyse de données, Modélisation et Apprentissage automatique [Grenoble]

In this paper, we introduce the MVSim architecture which is able to cluster multi-view datasets (i.e. datasets containing several objects linked together by di erent relations), by using several instances of a co-similarity algorithm.We show that this framework provides better results than existing approaches, while reducing both time and space complexities thanks to an e cient parallelization of the computations. This approach allows to split large datasets into a set of smaller ones.

Mots clés: Multiview Learning; Similarity Learning; Co-clustering

Domaines: Apprentissage [cs.LG]; Intelligence artificielle [cs.AI]

Gilles Bisson, Clément Grimal. An Architecture to Efficiently Learn Co-Similarities from Multi-View Datasets. International Conference on Neural Information Processing (ICONIP), Nov 2012, Doha, Qatar. pp.184-193. ⟨hal-00740749⟩