Publications
Students with bold-font names are primarily advised by me when doing the corresponding project.
Preprint
Book Chapter
Pan Li, Jure Leskovec, "The Expressive Power of Graph Neural Networks, " Chapter 5 in "Graph Neural Networks: Foundations, Frontiers, and Applications" edited by Wu, Lingfei and Cui, Peng and Pei, Jian and Zhao, Liang. (Also see the formal book link here: Graph Neural Networks: Foundations, Frontiers, and Applications | SpringerLink)
Eli Chien, Mufei Li, Anthony Aportela, Kerr Ding, Shuyi Jia, Supriyo Maji, Zhongyuan Zhao, Javier Duarte, Victor Fung, Callie Hao, Yunan Luo, Olgica Milenkovic, David Pan, Santiago Segarra, Pan Li, "Exploring the Opportunities and Challenges of Graph Neural Networks in Electrical Engineering," in Nature Reviews Electrical Engineering, 2024
Foundations of Machine Learning on Graphs
Siqi Miao, Zhiyuan Lu, Mia Liu, Javier Duarte, Pan Li, Locality-Sensitive Hashing-Based Efficient Point Transformer with Applications in High-Energy Physics, ICML 2024 (codes, oral)
Shikun Liu, Deyu Zou, Han Zhao, Pan Li, Pairwise Alignment Improves Graph Domain Adaptation, ICML 2024 (codes, spotlight)
Xiyuan Wang, Pan Li, Muhan Zhang, Graph As Point Set, ICML 2024
Rongzhe Wei, Eleonora Kreacic, Haoyu Wang, Haoteng Yin, Eli Chien, Vamsi K. Potluru, Pan Li, On the Inherent Privacy Properties of Discrete Denoising Diffusion Models. TMLR, WSDAIF 2023 Oral
Yinan Huang, William Lu, Joshua Robinson, Yu Yang, Muhan Zhang, Stefanie Jegelka, Pan Li, "On the Stability of Expressive Positional Encodings for Graphs, " ICLR 2024 (codes)
Peihao Wang, Shenghao Yang, Shu Li, Zhangyang Wang, Pan Li, "Polynomial Width is Sufficient for Set Representation with High-dimensional Features, " ICLR 2024
Tianci Liu, Haoyu Wang, Feijie Wu, Hengtong Zhang, Pan Li, Lu Su, Jing Gao, "Towards Poisoning Fair Representations, " ICLR 2024
Eli Chien, Wei-Ning Chen, Chao Pan, Pan Li, Ayfer Özgür, Olgica Milenkovic, "Differentially Private Decoupled Graph Convolutions for Multigranular Topology Protection, " NeurIPS 2023 code
Shikun Liu, Tianchun Li, Yongbin Feng, Nhan Tran, Han Zhao, Qiang Qiu, Pan Li, "Structural Re-weighting Improves Graph Domain Adaptation, " ICML 2023. (codes)
Siqi Miao, Yunan Luo, Mia Liu, Pan Li, "Interpretable Geometric Deep Learning via Learnable Randomness Injection," ICLR 2023. (codes)
Haoyu Wang, Pan Li, "Unsupervised Learning for Combinatorial Optimization Needs Meta-Learning," ICLR 2023. (codes)
Peihao Wang, Shenghao Yang, Yunyu Liu, Zhangyang Wang, Pan Li, "Equivariant Hypergraph Diffusion Neural Operators," ICLR 2023. (codes)
Rongzhe Wei, Haoteng Yin, Junteng Jia, Austin R. Benson, Pan Li, "Understanding Non-linearity in Graph Neural Networks from the Bayesian-Inference Perspective", NeurIPS 2022 (codes)
Haoyu Wang, Nan Wu, Hang Yang, Cong Hao, Pan Li, "Unsupervised Learning for Combinatorial Optimization with Principled Objective Relaxation," NeurIPS 2022 (codes)
Siqi Miao, Miaoyuan Liu, Pan Li, "Interpretable and Generalizable Graph Learning via Stochastic Attention Mechanism," ICML 2022 (codes)
Haorui Wang, Haoteng Yin, Muhan Zhang, Pan Li, "Equivariant and Stable Positional Encoding for More Powerful Graph Neural Networks," ICLR 2022. (code)
Mingyue Tang*, Carl Yang*, Pan Li, "Graph Auto-Encoder via Neighborhood Wasserstein Reconstruction," ICLR 2022 (code).
Muhan Zhang, Pan Li, Yinglong Xia, Kai Wang, Long Jin, "Labeling Trick: A Theory of Using Graph Neural Networks for Multi-Node Representation Learning," NeurIPS 2021.(code)
Muhan Zhang, Pan Li, "Nested Graph Neural Networks," NeurIPS 2021. (code)
Susheel Suresh, Pan Li, Cong Hao, Jennifer Neville, "Adversarial Graph Augmentation to Improve Graph Contrastive Learning," NeurIPS 2021. (codes)
Yanbang Wang, Yen-Yu Chang, Yunyu Liu, Jure Leskovec, Pan Li, "Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks," ICLR 2021. (codes)
Eli Chien*, Jianhao Peng*, Pan Li, Olgica Milenkovic, "Adaptive Universal Generalized PageRank Graph Neural Network, " ICLR 2021. (codes)
Pan Li, Yanbang Wang, Hongwei Wang, Jure Leskovec, "Distance Encoding -- Design Provably More Powerful GNNs for Structural Representation Learning," NeurIPS 2020. (codes)(slides)
Tailin Wu*, Hongyu Ren*, Pan Li, Jure Leskovec, "Graph Information Bottleneck," NeurIPS 2020. (codes) (slides)
Carl Yang, Peiye Zhuang, Wenhan Shi, Alan Luu, Pan Li, "Conditional Structure Generation through Graph Variational Generative Adversarial Nets," NeurIPS 2019.
Pan Li*, Eli Chien* and Olgica Milenkovic, "Optimizing Generalized PageRank Methods for Seed-Expansion Community Detection," NeurIPS 2019. (slides, poster, talk)
Eli Chien, Huozhi Zhou and Pan Li, "HS2: Active Learning on Hypergraphs," AISTATS 2019. (full-version)
Computation on Graphs
Siqi Miao, Zhiyuan Lu, Mia Liu, Javier Duarte, Pan Li, Locality-Sensitive Hashing-Based Efficient Point Transformer with Applications in High-Energy Physics, ICML 2024 (codes, recommended for oral)
Changlin Wan, Muhan Zhang, Pengtao Dang, Wei Hao, Sha Cao, Pan Li and Chi Zhang, "Ambiguities in Neural-Network-based Hyperedge Prediction", Journal of Applied and Computational Topology, 2024
Tianyi Zhang*, Haoteng Yin*, Rongzhe Wei, Pan Li, Anshumali Shrivastava, "Learning Scalable Structural Representations for Link Prediction with Bloom Signatures," WWW 2024 (codes)
Haoteng Yin, Muhan Zhang, Jianguo Wang, Pan Li, "SUREL+: Moving from Walks to Sets for Scalable Subgraph-based Graph Representation Learning," VLDB 2023 (codes)
Sushee Suresh, Mayank Shrivastava, Arko Mukherjee, Jennifer Neville, Pan Li, "Expressive and Efficient Representation Learning for Ranking Links in Temporal Graphs" WWW 2023 (codes)
Yuhong Luo, Pan Li, "Neighborhood-aware Scalable Temporal Network Representation Learning," LoG 2022 (best paper award!) (codes) (talks)
Haoyu Wang, Nan Wu, Hang Yang, Cong Hao, Pan Li, "Unsupervised Learning for Combinatorial Optimization with Principled Objective Relaxation," NeurIPS 2022 (codes)
Haoteng Yin, Muhan Zhang, Yanbang Wang, Jianguo Wang, Pan Li, "Algorithm and System Co-design for Efficient Subgraph-based Graph Representation Learning," VLDB 2022 (codes)
Muyi Liu, Pan Li, "SATMargin: Practical Maximal Frequent Subgraph Mining via Margin Space Sampling," WWW 2022
Kimon Fountoulakis, Pan Li, Shenghao Yang, "Local Hyper-flow Diffusion," NeurIPS 2021
Yuhong Li, Callie Hao, Pan Li, Jinjun Xiong, Deming Chen "Generic Neural Architecture Search via Regression, " NeurIPS 2021 (spotlight presentation, acceptance rate < 3%) (code)
Eli Chien, Pan Li, Olgica Milenkovic, "Landing Probabilities of Random Walks for Seed-set Expansion in Hypergraphs", ITW 2021.
Meng Liu, Nate Veldt, Haoyu Song, Pan Li, David Gleich, "Strongly Local Hypergraph Diffusions for Clustering and Semi-Supervised Learning", WWW 2021. (codes)
Lixiang Li, Yao Chen, Zacharie Zirnheld, Pan Li, and Cong Hao, "MeLoPPR: Software/Hardware Co-design for Memory-efficient Low-latency Personalized PageRank", DAC 2021. (codes)
Pan Li, Niao He and Olgica Milenkovic, "Quadratic Decomposable Submodular Function Minimization: Theory and Practice", JMLR 2020, NeurIPS 2018. (slides, slides (PhD defense) , codes, poster, neurips2018-version) --- Computation and Analysis of PageRanks over Hypergraphs!
Pan Li and Olgica Milenkovic, "Revisiting Decomposable Submodular Function Minimization with Incidence Relations," NeurIPS 2018. (slides (PhD defense) , codes, poster, full-version) --- Computation of min-cuts over Hypergraphs!
Pan Li and Olgica Milenkovic, "Submodular Hypergraph: p-Laplacian, Cheeger Inequalities and Spectral Clustering," ICML 2018. (slides, slides (PhD prelim), codes, poster, full-version) --- Spectral Theory for Hypergraphs!
Pan Li and Olgica Milenkovic, "Inhomogoenous Hypergraph Clustering with Applications," NeurIPS 2017 (spotlight presentation, acceptance rate < 4.6%). (slides, slides (PhD prelim), codes, poster, full-version) --- Applications motivate the line of research for Hypergraphs and preliminary spectral theory!
Pan Li, Gregory Puleo and Olgica Milenkovic, "Motif and Hypergraph Correlation Clustering," IEEE Transactions on Information Theory (TIT) 2019. (infocom2017-version, slides (phd qual))
Carl Yang, Yichen Feng, Pan Li, Yu Shi, and Jiawei Han, "Meta-Graph Based HIN Spectral Embedding: Methods, Analyses, and Insights," ICDM, 2018 (regular paper, acceptance rate < 8.9%).
Applications of Machine Learning over Graphs
Amit Roy, Juan Shu, Jia Li, Carl Yang, Olivier Elshocht, Jeroen Smeets, Pan Li, GAD-NR: Graph Anomaly Detection via Neighborhood Reconstruction, WSDM 2024 (code)
Tianchun Li*, Shikun Liu*, Yongbin Feng*, Garyfallia Paspalaki , Nhan V. Tran, Miaoyuan Liu , Pan Li, "Semi-supervised Graph Neural Network for Particle-level Noise Removal," The European Physics Journal C 2023, a short version appeared at NeurIPS AI4Science workshop 2021.
Guanhong Tao, Guangyu Shen, Yingqi Liu, Shengwei An, Qiuling Xu Shiqing Ma, Pan Li, Xiangyu Zhang, "Better Trigger Inversion Optimization in Backdoor Scanning," CVPR 2022 (oral) (codes)
Nan Wu, Hang Yang, Yuan Xie, Pan Li, Cong Hao, "High-Level Synthesis Performance Prediction using GNNs: Benchmarking, Modeling, and Advancing," DAC 2022
Yunyu Liu, Jianzhu Ma, Pan Li, "Neural Higher-order Pattern (Motif) Prediction in Temporal Networks," WWW 2022. (codes)
Houye Ji, Cheng Yang, Chuan Shi, Pan Li, Heterogeneous Graph Neural Network with Distance Encoding, ICDM 2021 and the full version in TKDE.
Hejie Cui*, Zijie Lu*, Pan Li, and Carl Yang, On Positional and Structural Node Features for Graph Neural Networks on Non-attributed Graphs, KDD DLG workshop 2021.
Susheel Suresh, Vinith Budde, Jennifer Neville, Pan Li, Jianzhu Ma, "Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns, " KDD 2021. (codes)
Andrew Z Wang, Rex Ying, Pan Li, Nikhil Rao, Karthik Subbian, Jure Leskovec, "Bipartite Dynamic Representations for Abuse Detection, " KDD 2021.
Yanbang Wang*, Pan Li*, Chongyang Bai, Jure Leskovec, "TEDIC: Neural Modeling of Behavioral Patterns in Dynamic Social Interaction Networks, " WWW 2021.
Siddharth Bhatia, Arjit Jain, Pan Li, Ritesh Kumar, Bryan Hooi, "MSTREAM: Fast Anomaly Detection in Multi-Aspect Streams," WWW 2021 (best paper runner-up).
Yen-Yu Chang*, Pan Li*, Rok Sosic, Mohamed Ibrahim, Marco Schweighauser and Jure Leskovec, "F-FADE: Frequency Factorization for Anomaly Detection in Edge Streams," selected for oral presentation, WSDM 2021. (selected for oral presentation) (codes)
Machine Learning for Ranking
Pan Li, Zhen Qin, Xuanhui Wang and Donald Metzler, "Combining Decision Trees and Neural Networks for Learning to Rank in Personal Search," KDD 2019 (applied data science track, selected for oral presentation, acceptance rate < 6.5%). (my intern project in Google) (short video, slides)
Pan Li and Olgica Milenkovic, "Efficient Ranking Aggregation with Lehmer Code," AISTATS 2017. (poster, full-version)
Pan Li and Olgica Milenkovic, "Multiclass Minmax Rank Aggregation," ISIT 2017. (slides)
Publications before coming to UIUC
Wenbo Ding, Yang Lu, Fang Yang, Wei Dai, Pan Li, Sicong Liu, Jian Song, "Spectrally Efficient CSI Acquisition for Power Line Communications: A Bayesian Compressive Sensing Perspective," in the IEEE Journal on Selected Areas in Communications, 2016.
Pan Li, Wei Dai, Huadong Meng and Xiqin Wang, "On Recovery of Sparse Signals with Block Structures," ISIT, 2015.
Pan Li, Huadong Meng, and Xiqin Wang, "A Feature Selection Method based on the Sparse MultiClass SVM for Fingerprinting Localization," VTC fall, 2014.
Chundi Zheng, Gang Li, Pan Li and XiqinWang, "Hyperparameter-free DOA Estimation under Power Constraints," ICASSP, 2013.
Selected Publications:
Neural modeling of network motifs: we proposed a novel neural encoding tool of network motifs, called causal anonymous walk, to inductively represent network dynamics. Casual anonymous walks automatically model and learn the impact of temporal network motifs on the evolvement of network structures. For example, casual anonymous walks can be used to model the triadic closure in social network evolving.
Power of Generalized PageRank in GNNs: Generalized PageRank solves every issue known till now in GNNs for node classification, including the over-smoothing issue, the overfitting issue, and the inapplicability to heterophilic networks, etc. Our results show that for node classification, GNNs do not need to add any non-linearity during message passing procedure. Different hops of message passing being associated with scalar learnable weights is good enough to capture the potentially complex structural relation in node classification tasks. This observation also demonstrates our previous theory by analyzing random-walk-type message passing over networks.
with Yanbang Wang, Hongwei Wang, Jure Leskovec
A follow-up work on practical explanations of how distance encoding helps graph representation learning. We also include more real-data experiments, especially on node classification over heterophilic networks:
with Haoteng Yin, Yanbang Wang
--- codes