Discovery and Separation of Features for Invariant Representation Learning

2 Dec 2019 · Ayush Jaiswal, Rob Brekelmans, Daniel Moyer, Greg Ver Steeg, Wael Abd-Almageed, Premkumar Natarajan ·

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.

PDF Abstract