Disentanglement and Generalization Under Correlation Shifts

Christina M. Funke, Paul Vicol, Kuan-Chieh Wang, Matthias Kümmerer, Richard S. Zemel, Matthias Bethge

Abstract: Correlations between factors of variation are prevalent in real-world data. Exploiting such correlations may increase predictive performance on noisy data; however, often correlations are not robus...