Forward Compatible Few-Shot Class-Incremental Learning

• Published in: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp. 9036 - 9046
• Main Authors: Zhou, Da-Wei, Wang, Fu-Yun, Ye, Han-Jia, Ma, Liang, Pu, Shiliang, Zhan, De-Chuan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2022
• Subjects: Computational modeling; Learning systems; Machine learning; Predictive models; Prototypes; Resists; Training; Transfer/low-shot/long-tail learning

Novel classes frequently arise in our dynamically changing world, e.g., new users in the authentication system, and a machine learning model should recognize new classes without forgetting old ones. This scenario becomes more challenging when new class instances are insufficient, which is called few...