Combat data shift in few-shot learning with knowledge graph

作     者：Yongchun ZHU Fuzhen ZHUANG Xiangliang ZHANG Zhiyuan QI Zhiping SHI Juan CAO Qing HE Yongchun ZHU;Fuzhen ZHUANG;Xiangliang ZHANG;Zhiyuan QI;Zhiping SHI;Juan CAO;Qing HE

作者机构：Key Lab of Intelligent Information Processing of Chinese Academy of Sciences(CAS)Institute of Computing TechnologyCASBeijing 100190China University of Chinese Academy of SciencesBeijing 100049China Institute of Artificial IntelligenceBeihang UniversityBeijing 100191China Xiamen Institute of Data IntelligenceXiamen 361021China Computer Science and EngineeringUniversity of Notre DameIN 46556USA University of California San DiegoLa JollaCA 92093USA College of Information Engineering&Academy for Multidisciplinary StudiesCapital Normal UniversityBeijing 100089China 

出 版 物：《Frontiers of Computer Science》 (中国计算机科学前沿（英文版）)

年 卷 期：2023年第17卷第1期

页      面：101-111页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 

基　　金：supported by the National Natural Science Foundation of China (Grant Nos. 62176014  U1836206  61773361  U1811461) 

主　　题：few-shot data shift knowledge graph 

摘      要：Many few-shot learning approaches have been designed under the meta-learning framework, which learns from a variety of learning tasks and generalizes to new tasks. These meta-learning approaches achieve the expected performance in the scenario where all samples are drawn from the same distributions (i.i.d. observations). However, in real-world applications, few-shot learning paradigm often suffers from data shift, i.e., samples in different tasks, even in the same task, could be drawn from various data distributions. Most existing few-shot learning approaches are not designed with the consideration of data shift, and thus show downgraded performance when data distribution shifts. However, it is non-trivial to address the data shift problem in few-shot learning, due to the limited number of labeled samples in each task. Targeting at addressing this problem, we propose a novel metric-based meta-learning framework to extract task-specific representations and task-shared representations with the help of knowledge graph. The data shift within/between tasks can thus be combated by the combination of task-shared and task-specific representations. The proposed model is evaluated on popular benchmarks and two constructed new challenging datasets. The evaluation results demonstrate its remarkable performance.