Wael AbdAlmageed

Discovery and Separation of Features for Invariant Representation Learning

TitleDiscovery and Separation of Features for Invariant Representation Learning
Publication TypeJournal Article
Year of Publication2019
AuthorsA. Jaiswal, R. Brekelmans, D. Moyer, G. Ver Steeg, W. AbdAlmageed, and P. Natarajan
JournalarXiv:1912.00646 [cs, stat]
Abstract

Supervised machine learning models often associate irrelevant nuisance factors with the prediction target, which hurts generalization. We propose a framework for training robust neural networks that induces invariance to nuisances through learning to discover and separate predictive and nuisance factors of data. We present an information theoretic formulation of our approach, from which we derive training objectives and its connections with previous methods. Empirical results on a wide array of datasets show that the proposed framework achieves state-of-the-art performance, without requiring nuisance annotations during training.

URLhttp://arxiv.org/abs/1912.00646
Groups: