Learn more from less: Generalized zero-shot learning with severely limited labeled data

Neurocomputing(2022)

引用 4|浏览14
暂无评分
摘要
•We highlight the FS-GZSL problem, and two data scarcity settings are proposed.•A novel framework MAML-CLA is proposed to address FS-GZSL by mimicking ZSL setting.•MAML-CLA outperforms state-of-the-art methods on popular datasets under two setting.
更多
查看译文
关键词
Generalized zero-shot learning,Data scarcity scenario,Meta-learning paradigm,Entropy gate
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要