Match!
Tat-Seng Chua
National University of Singapore
Machine learningData miningPattern recognitionInformation retrievalComputer science
601Publications
71H-index
19.9kCitations
What is this?
Publications 572
Newest
#1Zhulin Tao (CUC: Communication University of China)H-Index: 1
#2Yinwei WeiH-Index: 3
Last. Tat-Seng Chua (NUS: National University of Singapore)H-Index: 71
view all 6 authors...
Abstract Graph neural networks (GNNs) have shown great potential for personalized recommendation. At the core is to reorganize interaction data as a user-item bipartite graph and exploit high-order connectivity among user and item nodes to enrich their representations. While achieving great success, most existing works consider interaction graph based only on ID information, foregoing item contents from multiple modalities (e.g., visual, acoustic, and textual features of micro-video items). Dist...
Source
Jul 25, 2020 in SIGIR (International ACM SIGIR Conference on Research and Development in Information Retrieval)
#1Xun Yang (NUS: National University of Singapore)H-Index: 8
#2Jianfeng Dong (ZJSU: Zhejiang Gongshang University)
Last. Tat-Seng Chua (NUS: National University of Singapore)H-Index: 71
view all 6 authors...
Jul 25, 2020 in SIGIR (International ACM SIGIR Conference on Research and Development in Information Retrieval)
#1Chen Qian (THU: Tsinghua University)
#2Fuli Feng (NUS: National University of Singapore)H-Index: 11
Last. Tat-Seng Chua (NUS: National University of Singapore)H-Index: 71
view all 5 authors...
Jul 25, 2020 in SIGIR (International ACM SIGIR Conference on Research and Development in Information Retrieval)
#1Xingchen Li (ZJU: Zhejiang University)
#2Xiang Wang (NUS: National University of Singapore)H-Index: 55
Last. Tat-Seng Chua (NUS: National University of Singapore)H-Index: 71
view all 6 authors...
Jul 25, 2020 in SIGIR (International ACM SIGIR Conference on Research and Development in Information Retrieval)
#1Xiang Wang (NUS: National University of Singapore)H-Index: 55
#2Hongye Jin (PKU: Peking University)
Last. Tat-Seng Chua (NUS: National University of Singapore)H-Index: 71
view all 6 authors...
Jul 5, 2020 in ACL (Meeting of the Association for Computational Linguistics)
#1Yixin Cao (NUS: National University of Singapore)H-Index: 8
#1Yixin Cao (NUS: National University of Singapore)H-Index: 5
Last. Tat-Seng Chua (NUS: National University of Singapore)H-Index: 71
view all 6 authors...
Jul 5, 2020 in ACL (Meeting of the Association for Computational Linguistics)
#1Liangming Pan (NUS: National University of Singapore)H-Index: 3
Last. Min-Yen Kan (SF: Salesforce.com)H-Index: 41
view all 5 authors...
#2Tat-Seng ChuaH-Index: 71
Last. Gim Hee Lee (NUS: National University of Singapore)H-Index: 23
view all 3 authors...
Many existing approaches for point cloud semantic segmentation are strongly supervised. These strongly supervised approaches heavily rely on a large amount of labeled training data that is difficult to obtain and suffer from poor generalization to new classes. To mitigate these limitations, we propose a novel attention-aware multi-prototype transductive few-shot point cloud semantic segmentation method to segment new classes given a few labeled examples. Specifically, each class is represented b...
#1Francesco Gelli (NUS: National University of Singapore)H-Index: 5
#2Tiberio Uricchio (UniFI: University of Florence)H-Index: 11
Last. Tat-Seng Chua (NUS: National University of Singapore)H-Index: 71
view all 5 authors...
Source
Jun 14, 2020 in CVPR (Computer Vision and Pattern Recognition)
#2Jingjing Chen (Fudan University)H-Index: 6
Last. Yu-Gang Jiang (Fudan University)H-Index: 41
view all 6 authors...
This paper proposes a Hyperbolic Visual Embedding Learning Network for zero-shot recognition. The network learns image embeddings in hyperbolic space, which is capable of preserving the hierarchical structure of semantic classes in low dimensions. Comparing with existing zero-shot learning approaches, the network is more robust because the embedding feature in hyperbolic space better represents class hierarchy and thereby avoid misleading resulted from unrelated siblings. Our network outperforms...
12345678910