I’m curious - does anyone here have experience with mixup helping training in problems with severe class imbalance? The original mixup paper mentions it obliquely but doesn’t directly address it (“More related to mixup, Chawla et al. (2002) propose to augment the rare class in an imbalanced dataset by interpolating the nearest neighbors; DeVries & Taylor (2017) show that interpolation and extrapolation the nearest neighbors of the same class in feature space can improve generalization.”).
3 Likes