.png){kind=icon}
返回
Semantic alignment with self-supervision for class incremental learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.knosys.2023.111114.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Existing class incremental learning methods typically employ knowledge distillation to minimize discrepancies in model outputs. However, these methods are restricted by the mismatch between quondam knowledge and new data. To alleviate these issues, we introduce semantic alignment decouples the classification and distillation in different semantic spaces. The unmatched new data is regarded as out-of-distribution data on the old class distribution, and the corresponding pseudo-labels are attached to the new data using the original network. Intuitively, the pseudo-labels could be consistently preserved in the old semantic space. Moreover, we develop auxiliary self-supervised classifiers to learn more generalized representation, enabling a better stability plastic trade-off. Furthermore, self-distillation is employed to refine self-supervised knowledge from auxiliary classifiers. Extensive experiments demonstrate that our method achieves the best performance on CIFAR100, ImageNet100, ImageNet, CUB200, and Stanford-Dogs120 datasets. Notably, our method outperforms existing methods by a substantial margin when only one old exemplar is stored per class, i.e., 11.34% and 21.46% improvement on CIFAR100 of 5 phases and 10 phases, respectively.
Keyword:
Class incremental learning
Semantic alignment
Self-supervision
Self-distillation
期刊
K
IF:
7.6
论文数:
1.2W
被引数:
4.5W
机构
暂无机构信息
引用论文
Phase II Multicenter Trial of Lenalidomide: Clinical and Immunomodulatory Effects in Patients with CTCL.png)
Blood
BloodIF0
PREAI
分享收藏
Developing Wind and/or Solar Powered Crop Irrigation Systems for the Great Plains为大平原开发风能和/或太阳能作物灌溉系统
PREAI
分享收藏Multi-feature space similarity supplement for few-shot class incremental learning少镜头类增量学习的多特征空间相似性补充
PREAI
分享收藏 没有更多内容
没有更多内容