106 citations
- Meta (United States)US2 papers
- Shenzhen Institutes of Advanced TechnologyCN2 papers
- University of Maryland, College ParkUS2 papers
- Aalto UniversityFI1 paper
- China University of GeosciencesCN1 paper
- Chinese Academy of SciencesCN1 paper
- Chinese University of Hong Kong, ShenzhenCN1 paper
- Commonwealth Scientific and Industrial Research OrganisationAU1 paper
- Data61AU1 paper
- GE Global Research (United States)US1 paper
- General Electric (United States)US1 paper
- Huazhong University of Science and TechnologyCN1 paper
Showing 2019 · cs.LGShow all
2 papers · 2 filters
cs.LG2019★ 14 cited
Cross-Batch Memory for Embedding Learning
Xun Wang, Haozhi Zhang, Weilin Huang +1
Mining informative negative instances are of central importance to deep metric learning (DML), however this task is intrinsically limited by mini-batch training, where only a mini-…
cs.LG2019★ 2 cited
HAL: Improved Text-Image Matching by Mitigating Visual Semantic Hubs
Fangyu Liu, Rongtian Ye, Xun Wang +1
The hubness problem widely exists in high-dimensional embedding space and is a fundamental source of error for cross-modal matching tasks. In this work, we study the emergence of h…