To classify in-distribution samples, deep neural networks explore strongly label-related information and discard weakly label-related information according to the information bottleneck. Out-of-distribution samples drawn from distributions differing from that of in-distribution samples could be assigned with unexpected high-confidence predictions because they could obtain minimum strongly label-related information. To distinguish in- and out-of-distribution samples, Dual Representation Learning (DRL) makes out-of-distribution samples harder to have high-confidence predictions by exploring both strongly and weakly label-related information from in-distribution samples. For a pretrained network exploring strongly label-related information to learn label-discriminative representations, DRL trains its auxiliary network exploring the remaining weakly label-related information to learn distribution-discriminative representations. Specifically, for a label-discriminative representation, DRL constructs its complementary distribution-discriminative representation by integrating diverse representations less similar to the label-discriminative representation. Accordingly, DRL combines label- and distribution-discriminative representations to detect out-of-distribution samples. Experiments show that DRL outperforms the state-of-the-art methods for out-of-distribution detection.
Deep Dual Support Vector Data Description for Anomaly Detection on Attributed Networks
Zhengzhou University(郑州大学);GRG Banking Technology Co., Ltd(广电运通金融电子股份有限公司);Harbin University of Science and Technology(哈尔滨理工大学);Minjiang University(闽江学院)
Graph-Based Fraud Detection with Dual-Path Graph Filtering
School of Intelligent Systems Science and Engineering(智能系统科学与工程学院);Jinan University(暨南大学);Zhuhai 519070, China(珠海519070, 中国);Department of Computer Science(计算机科学系);University of Illinois Chicago(伊利诺伊大学芝加哥分校);Chicago 60607, USA(芝加哥60607, 美国)
Dual-view Aware Smart Contract Vulnerability Detection for Ethereum
Institute of Cyberspace Security, Zhejiang University of Technology(浙江工业大学网络空间安全研究院);Binjiang Institute of Artificial Intelligence, ZJUT(浙江工业大学滨江人工智能研究院)
ESTM: An Enhanced Dual-Branch Spectral-Temporal Mamba for Anomalous Sound Detection
Shenzhen International Graduate School, Tsinghua University(清华大学深圳国际研究生学院);Collaborative Innovation Center for Transport Studies, Dalian Maritime University(大连海运大学交通运输协同创新中心)
MTDA-HSED: Mutual-Assistance Tuning and Dual-Branch Aggregating for Heterogeneous Sound Event Detection
School of Artificial Intelligence, Beijing University of Posts and Telecommunications(北京邮电大学人工智能学院);School of Intelligent Engineering and Automation, Beijing University of Posts and Telecommunications(北京邮电大学智能工程与自动化学院)
WiFi CSI Based Temporal Activity Detection via Dual Pyramid Network
School of Information and Communication Engineering, University of Electronic Science and Technology of China(电子科技大学信息与通信工程学院);College of Computer Science, Sichuan University(四川大学计算机学院)