Knowledge Commons of Institute of Automation,CAS
Class-Incremental Learning via Dual Augmentation | |
Zhu Fei (朱飞); Zhen Cheng; Xu-Yao Zhang; Cheng-Lin Liu | |
2021-12-06 | |
会议名称 | Thirty-fifth Conference on Neural Information Processing Systems |
会议日期 | Dec 6-14, 2021 |
会议地点 | Virtual |
出版者 | MIT Press |
摘要 | Deep learning systems typically suffer from catastrophic forgetting of past knowledge when acquiring new skills continually. In this paper, we emphasize two dilemmas, representation bias and classifier bias in class-incremental learning, and present a simple and novel approach that employs explicit class augmentation (classAug) and implicit semantic augmentation (semanAug) to address the two biases, respectively. On the one hand, we propose to address the representation bias by learning transferable and diverse representations. Specifically, we investigate the feature representations in incremental learning based on spectral analysis and present a simple technique called classAug, to let the model see more classes during training for learning representations transferable across classes. On the other hand, to overcome the classifier bias, semanAug implicitly involves the simultaneous generating of an infinite number of instances of old classes in the deep feature space, which poses tighter constraints to maintain the decision boundary of previously learned classes. Without storing any old samples, our method can perform comparably with representative data replay based approaches. |
七大方向——子方向分类 | 模式识别基础 |
国重实验室规划方向分类 | 人工智能基础前沿理论 |
是否有论文关联数据集需要存交 | 否 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/52408 |
专题 | 多模态人工智能系统全国重点实验室_模式分析与学习 |
作者单位 | 1.NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China 2.University of Chinese Academy of Sciences, Beijing, 100049, China |
推荐引用方式 GB/T 7714 | Zhu Fei ,Zhen Cheng,Xu-Yao Zhang,et al. Class-Incremental Learning via Dual Augmentation[C]:MIT Press,2021. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
NeurIPS-2021-class-i(1415KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论