Class-Incremental Learning via Dual Augmentation Venue: NeurIPS 2021 Problem: Representation Bias and Classifier Bias Method: Class Aug和 Semantic Aug SemanticAug: 作者假设每个类在deep feature space服从高斯分布,将历史数据的均值和方差记录下来。在对模型做更新时,作者用这些分布信息生成一些“语义特征”喂给...
Class-incremental learning (CIL) aims to recognize new classes incrementally while maintaining the discriminability of old classes. Most existing CIL methods are exemplar-based, i.e., storing a part of old data for retraining. Without relearning old data, those methods suffer from catastrophic forget...
class confusion.To tackle this issue, we propose a simple yet effective method through cross-domain concePt INtegrAtion (PINA).We train a Unified Classifier (UC) as a concept container across all domains.Then, a Domain Specific Alignment (DSA) module is proposed for each incremental domain, ...