Learning Adaptive Embedding Considering Incremental Class
Abstract
Class-Incremental Learning (CIL) aims to train a reliable model with the streaming data, which emerges unknown classes sequentially. Different from traditional closed set learning, CIL has two main challenges: (1) Novel class detection. The initial training data only contains incomplete classes, and streaming test data will accept unknown classes. Therefore, the model needs to not only accurately classify known classes, but also effectively detect unknown classes; (2) Model expansion. After the novel classes are detected, the model needs to be updated without re-training using the entire previous data. However, traditional CIL methods have not fully considered these two challenges. To this end, we propose a Class-Incremental Learning without Forgetting (CILF) framework. In detail, CILF designs to regularize classification with decoupled prototype based loss, which can improve the intra-class and inter-class structure significantly, and acquire a compact embedding representation for novel class detection in result. Then, CILF employs a learnable curriculum clustering operator to estimate the number of semantic clusters via fine-tuning the learned network, in which curriculum operator can adaptively learn the embedding in self-taught form. Last, with the labeled streaming test data, CILF can update the network with robust regularization to mitigate the catastrophic forgetting.
Recommended Citation
Y. Yang et al., "Learning Adaptive Embedding Considering Incremental Class," IEEE Transactions on Knowledge and Data Engineering, Institute of Electrical and Electronics Engineers (IEEE), Sep 2021.
The definitive version is available at https://doi.org/10.1109/TKDE.2021.3109131
Department(s)
Computer Science
Publication Status
Early Access
Keywords and Phrases
Class-Incremental Learning; Data Models; Incremental Model Update; Labeling; Manuals; Novel Class Detection; Prototypes; Streaming Media; Task Analysis; Training
International Standard Serial Number (ISSN)
1558-2191; 1041-4347
Document Type
Article - Journal
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2021 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.
Publication Date
02 Sep 2021
Comments
This research was supported by NSFC (62006118,61836013,91746301), Natural Science Foundation of Jiangsu Province of China under Grant (BK20200460), CCF-Baidu Open Fund (CCF-BAIDU OF2020011), Baidu Fund (20121202OT03645).