OvercomingCatastrophicForgettinginIncremental Few-ShotLearningbyFindingFlatMinima

This paper considers incremental few-shot learning, which requires a model to continually recognize new categories with only a few examples provided. Our study shows that existing methods severely suffer from catastrophic forgetting, awell-known problem in incremental learning, which is aggravated due to data scarcity andimbalance inthefew-shot setting.