Professor
School of Data Science
School of Cyber Science and Technology
School of Information Science and Technology
University of Science and Technology of China
Email: xiangwang1223 at gmail.com
• Google Scholar Page • GitHub Page
Biography
I am a Professor in University of Science and Technology of China, where I am a member of Lab of Data Science. With my colleagues, students, and collaborators, we strive to develop trustworthy deep learning and artificial intelligence algorithms with better interpretability, generalization, and robustness. Our research is motivated by, and contributes to, graph-structured applications in information retrieval (e.g., personalized recommendation), data mining (e.g., graph pre-training), security (e.g., fraud detection in fintech, information security in system), and multimedia (e.g., video question answering). Our work has over 50 publications in top-tier conferences and journals. Over 10 papers have been featured in the most cited and influential list (e.g., KDD 2019, SIGIR 2019, SIGIR 2020, SIGIR 2021) and best paper finalist (e.g., WWW 2021, CVPR 2022). Moreover, I have served as the PC member for top-tier conferences including NeurIPS, ICLR, SIGIR and KDD, and the invited reviewer for prestigious journals including JMLR, TKDE, and TOIS.
Prospective Ph.D., Master, and Undergraduate Students
I am looking for highly motivated students (PhD, master, undergraduate students) to work together on trustworthy deep learning on graph, especially pre-training, interpretability, generalization, and robustness, and their applications in real-world scenarios. Please feel free to send me your CV and transcripts, if you have interest. We are also actively looking for opportunities in research, partnership and commercialization in exciting data science projects.
News
Invariant Collaborative Filtering to Popularity Distribution Shift.
GIF: A General Graph Unlearning Strategy via Influence Function.
Addressing Heterophily in Graph Anomaly Detection: A Perspective of Graph Spectrum.
Boosting Causal Discovery via Adaptive Sample Reweighting.
Learning Graph-based Code Representations for Source-level Functional Similarity Detection.
Causal Inference for Knowledge Graph based Recommendation.
Alleviating Structural Distribution Shift in Graph Anomaly Detection.
Cooperative Explanations of Graph Neural Networks.
Incorporating Bias-aware Margins into Contrastive Loss for Collaborative Filtering.
User Perception of Recommendation Explanation: Are Your Explanations What Users Need?.
Equivariant and Invariant Grounding for Video Question Answering.
Let invariant Rationale Discovery inspire Graph Contrastive Learning.
Causal Attention for Interpretable and Generalizable Graph Classification.
CrossCBR: Cross-view Contrastive Learning for Bundle Recommendation.
Reinforced Causal Explainer for Graph Neural Networks.
Time-aware Path Reasoning on Knowledge Graph for Recommendation.
TELL: Log Level Suggestions via Modeling Multi-level Code Block Information.
Invariant Grounding for Video Question Answering.[Oral Presentation & Best Paper Finalist]
Temporal Feature Alignment and Mutual Information Maximization for VideoBased Human Pose Estimation.[Oral]
ShadeWatcher: Recommendation-guided Cyber Threat Analysis using System Audit Records.
Discovering invariant rationales for graph neural networks.
Highlights
Honors and Awards
Background
Supervisor: Prof Tat-Seng Chua
Supervisor: Prof Tat-Seng Chua
Supervisor: Prof Tat-Seng Chua; Mentor: Prof Xiangnan He, Prof Liqiang Nie