PAC-Bayesian Analysis for a two-step Hierarchical Multiview Learning Approach

Anil Goyal (1, 2) , Emilie Morvant (1) , Pascal Germain (3) , Massih-Reza Amini (2)

1 LHC - Laboratoire Hubert Curien
2 AMA - Analyse de données, Modélisation et Apprentissage automatique [Grenoble]
3 SIERRA - Statistical Machine Learning and Parsimony

We study a two-level multiview learning with more than two views under the PAC-Bayesian framework. This approach, sometimes referred as late fusion, consists in learning sequentially multiple view-specific classifiers at the first level, and then combining these view-specific classifiers at the second level. Our main theoretical result is a generalization bound on the risk of the majority vote which exhibits a term of diversity in the predictions of the view-specific classifiers. From this result it comes out that controlling the trade-off between diversity and accuracy is a key element for multiview learning, which complements other results in multiview learning. Finally, we experiment our principle on multiview datasets extracted from the Reuters RCV1/RCV2 collection.

Mots clés: Multiview Learning; PAC-Bayesian Theory

Domaines: Machine Learning [stat.ML]

Anil Goyal, Emilie Morvant, Pascal Germain, Massih-Reza Amini. PAC-Bayesian Analysis for a two-step Hierarchical Multiview Learning Approach. European Conference on Machine Learning & Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), Sep 2017, Skopje, Macedonia. ⟨hal-01546109⟩