ICLR2022图神经网络论文集锦

嘿，记得给“机器学习与推荐算法”添加星标

---

| 整理：潘润琦

| 链接：zhuanlan.zhihu.com/p/464093894

| 转自：深度学习与图网络

今年的ICLR录取结果出了，图神经网络也是今年的一大热点，这里总结一部分看到的GNN的文章，如果有错误的或者遗漏的文章请大家一定指出来。

https://github.com/naganandy/graph-based-deep-learning-literature/blob/master/conference-publications/folders/publications_iclr22/README.md

今年iclr感觉是GNN应用爆发的一年~

## Baseline and Benchmark

Graph-less Neural Networks: Teaching Old MLPs New Tricks Via Distillation Shichang Zhang, Yozen Liu, Yizhou Sun, Neil Shah

Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions Leslie O'Bray, Max Horn, Bastian Rieck, Karsten Borgwardt

GNN is a Counter? Revisiting GNN for Question Answering Kuan Wang, Yuyu Zhang, Diyi Yang, Le Song, Tao Qin

Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework Xu Ma, Can Qin, Haoxuan You, Haoxi Ran, Yun Fu

## GNN理论

A New Perspective on "How Graph Neural Networks Go Beyond Weisfeiler-Lehman?" Asiri Wijesinghe, Qing Wang

Frame Averaging for Invariant and Equivariant Network Design Omri Puny, Matan Atzmon, Edward J. Smith, Ishan Misra, Aditya Grover, Heli Ben-Hamu, Yaron Lipman

Equivariant Subgraph Aggregation Networks Beatrice Bevilacqua, Fabrizio Frasca, Derek Lim, Balasubramaniam Srinivasan, Chen Cai, Gopinath Balamurugan, Michael M. Bronstein, Haggai Maron

Understanding over-squashing and bottlenecks on graphs via curvature Jake Topping, Francesco Di Giovanni, Benjamin Paul Chamberlain, Xiaowen Dong, Michael M. Bronstein

Expressiveness and Approximation Properties of Graph Neural Networks Floris Geerts, Juan L Reutter

How Attentive are Graph Attention Networks? Shaked Brody, Uri Alon, Eran Yahav

From Stars to Subgraphs: Uplifting Any GNN with Local Structure Awareness Lingxiao Zhao, Wei Jin, Leman Akoglu, Neil Shah

Is Homophily a Necessity for Graph Neural Networks? Yao Ma, Xiaorui Liu, Neil Shah, Jiliang Tang

PF-GNN: Differentiable particle filtering based approximation of universal graph representations Mohammed Haroon Dupty, Yanfei Dong, Wee Sun Lee

Do We Need Anisotropic Graph Neural Networks? Shyam A. Tailor, Felix Opolka, Pietro Lio, Nicholas Donald Lane

LEARNING GUARANTEES FOR GRAPH CONVOLUTIONAL NETWORKS ON THE STOCHASTIC BLOCK MODEL Wei Lu

Triangle and Four Cycle Counting with Predictions in Graph Streams Justin Y Chen, Talya Eden, Piotr Indyk, Honghao Lin, Shyam Narayanan, Ronitt Rubinfeld, Sandeep Silwal, Tal Wagner, David Woodruff, Michael Zhang

Equivariant and Stable Positional Encoding for More Powerful Graph Neural Networks Haorui Wang, Haoteng Yin, Muhan Zhang, Pan Li

Why Propagate Alone? Parallel Use of Labels and Features on Graphs Yangkun Wang, Jiarui Jin, Weinan Zhang, Yang Yongyi, Jiuhai Chen, Quan Gan, Yong Yu, Zheng Zhang, Zengfeng Huang, David Wipf

You are AllSet: A Multiset Function Framework for Hypergraph Neural Networks Eli Chien, Chao Pan, Jianhao Peng, Olgica Milenkovic

Revisiting Over-smoothing in BERT from the Perspective of Graph Han Shi, JIAHUI GAO, Hang Xu, Xiaodan Liang, Zhenguo Li, Lingpeng Kong, Stephen M. S. Lee, James Kwok

The Role of Permutation Invariance in Linear Mode Connectivity of Neural Networks Rahim Entezari, Hanie Sedghi, Olga Saukh, Behnam Neyshabur

## 新的GNN结构

Convergent Graph Solvers Junyoung Park, Jinhyun Choo, Jinkyoo Park

Equivariant Subgraph Aggregation Networks Beatrice Bevilacqua, Fabrizio Frasca, Derek Lim, Balasubramaniam Srinivasan, Chen Cai, Gopinath Balamurugan, Michael M. Bronstein, Haggai Maron

Neural Structured Prediction for Inductive Node Classification Meng Qu, Huiyu Cai, Jian Tang

Geometric and Physical Quantities improve E(3) Equivariant Message Passing Johannes Brandstetter, Rob Hesselink, Elise van der Pol, Erik J Bekkers, Max Welling

Topological Graph Neural Networks Max Horn, Edward De Brouwer, Michael Moor, Yves Moreau, Bastian Rieck, Karsten Borgwardt

GRAND++: Graph Neural Diffusion with A Source Term Matthew Thorpe, Tan Minh Nguyen, Hedi Xia, Thomas Strohmer, Andrea Bertozzi, Stanley Osher, Bao Wang

Graph Neural Networks with Learnable Structural and Positional Representations Vijay Prakash Dwivedi, Anh Tuan Luu, Thomas Laurent, Yoshua Bengio, Xavier Bresson

## Deep GNN

Towards Deepening Graph Neural Networks: A GNTK-based Optimization Perspective Wei Huang, Yayong Li, weitao Du, Richard Xu, Jie Yin, Ling Chen, Miao Zhang

## Data Augmentation

Graph Condensation for Graph Neural Networks Wei Jin, Lingxiao Zhao, Shichang Zhang, Yozen Liu, Jiliang Tang, Neil Shah

## GNN训练

PipeGCN: Efficient Full-Graph Training of Graph Convolutional Networks with Pipelined Feature Communication Cheng Wan, Youjie Li, Cameron R. Wolfe, Anastasios Kyrillidis, Nam Sung Kim, Yingyan Lin

IGLU: Efficient GCN Training via Lazy Updates S Deepak Narayanan, Aditya Sinha, Prateek Jain, Purushottam Kar, SUNDARARAJAN SELLAMANICKAM

Learning to Schedule Learning rate with Graph Neural Networks Yuanhao Xiong, Li-Cheng Lan, Xiangning Chen, Ruochen Wang, Cho-Jui Hsieh

Towards Training Billion Parameter Graph Neural Networks for Atomic Simulations Anuroop Sriram, Abhishek Das, Brandon M Wood, C. Lawrence Zitnick

EXACT: Scalable Graph Neural Networks Training via Extreme Activation Compression Zirui Liu, Kaixiong Zhou, Fan Yang, Li Li, Rui Chen, Xia Hu

Learn Locally, Correct Globally: A Distributed Algorithm for Training Graph Neural Networks Morteza Ramezani, Weilin Cong, Mehrdad Mahdavi, Mahmut Kandemir, Anand Sivasubramaniam

Large-Scale Representation Learning on Graphs via Bootstrapping Shantanu Thakoor, Corentin Tallec, Mohammad Gheshlaghi Azar, Mehdi Azabou, Eva L Dyer, Remi Munos, Petar Veličković, Michal Valko

## 自监督

Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysis Siyi Tang, Jared Dunnmon, Khaled Kamal Saab, Xuan Zhang, Qianying Huang, Florian Dubost, Daniel Rubin, Christopher Lee-Messer

Pre-training Molecular Graph Representation with 3D Geometry Shengchao Liu, Hanchen Wang, Weiyang Liu, Joan Lasenby, Hongyu Guo, Jian Tang

Node Feature Extraction by Self-Supervised Multi-scale Neighborhood Prediction Eli Chien, Wei-Cheng Chang, Cho-Jui Hsieh, Hsiang-Fu Yu, Jiong Zhang, Olgica Milenkovic, Inderjit S Dhillon

Automated Self-Supervised Learning for Graphs Wei Jin, Xiaorui Liu, Xiangyu Zhao, Yao Ma, Neil Shah, Jiliang Tang

## 可解释性&因果

Discovering Invariant Rationales for Graph Neural Networks Yingxin Wu, Xiang Wang, An Zhang, Xiangnan He, Tat-Seng Chua

Explanations of Black-Box Models based on Directional Feature Interactions Aria Masoomi, Davin Hill, Zhonghui Xu, Craig P Hersh, Edwin K. Silverman, Peter J. Castaldi, Stratis Ioannidis, Jennifer Dy

DEGREE: Decomposition Based Explanation for Graph Neural Networks Qizhang Feng, Ninghao Liu, Fan Yang, Ruixiang Tang, Mengnan Du, Xia Hu

A Biologically Interpretable Graph Convolutional Network to Link Genetic Risk Pathways and Imaging Phenotypes of Disease Sayan Ghosal, Qiang Chen, Giulio Pergola, Aaron L Goldman, William Ulrich, Daniel R Weinberger, Archana Venkataraman

Explainable GNN-Based Models over Knowledge Graphs David Jaime Tena Cucala, Bernardo Cuenca Grau, Egor V. Kostylev, Boris Motik

Efficient Neural Causal Discovery without Acyclicity Constraints Phillip Lippe, Taco Cohen, Efstratios Gavves

## 图生成

Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design Wenhao Gao, Rocío Mercado, Connor W. Coley

Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-design Wengong Jin, Jeremy Wohlwend, Regina Barzilay, Tommi S. Jaakkola

Data-Efficient Graph Grammar Learning for Molecular Generation Minghao Guo, Veronika Thost, Beichen Li, Payel Das, Jie Chen, Wojciech Matusik

Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions Leslie O'Bray, Max Horn, Bastian Rieck, Karsten Borgwardt

GeoDiff: A Geometric Diffusion Model for Molecular Conformation Generation Minkai Xu, Lantao Yu, Yang Song, Chence Shi, Stefano Ermon, Jian Tang

Graph Auto-Encoder via Neighborhood Wasserstein Reconstruction Mingyue Tang, Pan Li, Carl Yang

GraphENS: Neighbor-Aware Ego Network Synthesis for Class-Imbalanced Node Classification Joonhyung Park, Jaeyun Song, Eunho Yang

Top-N: Equivariant Set and Graph Generation without Exchangeability Clement Vignac, Pascal Frossard

Differentiable Scaffolding Tree for Molecule Optimization Tianfan Fu, Wenhao Gao, Cao Xiao, Jacob Yasonik, Connor W. Coley, Jimeng Sun

Learning to Extend Molecular Scaffolds with Structural Motifs Krzysztof Maziarz, Henry Richard Jackson-Flux, Pashmina Cameron, Finton Sirockin, Nadine Schneider, Nikolaus Stiefl, Marwin Segler, Marc Brockschmidt

Vitruvion: A Generative Model of Parametric CAD Sketches Ari Seff, Wenda Zhou, Nick Richardson, Ryan P Adams

## 组合优化

Graph Neural Network Guided Local Search for the Traveling Salesperson Problem Benjamin Hudson, Qingbiao Li, Matthew Malencia, Amanda Prorok

Neural Models for Output-Space Invariance in Combinatorial Problems Yatin Nandwani, Vidit Jain, Mausam ., Parag Singla

What’s Wrong with Deep Learning in Tree Search for Combinatorial Optimization Maximilian Böther, Otto Kißig, Martin Taraz, Sarel Cohen, Karen Seidel, Tobias Friedrich

Learning Scenario Representation for Solving Two-stage Stochastic Integer Programs Yaoxin Wu, Wen Song, Zhiguang Cao, Jie Zhang

## 物理

Constrained Graph Mechanics Networks Wenbing Huang, Jiaqi Han, Yu Rong, Tingyang Xu, Fuchun Sun, Junzhou Huang

Learning the Dynamics of Physical Systems from Sparse Observations with Finite Element Networks Marten Lienen, Stephan Günnemann

R5: Rule Discovery with Reinforced and Recurrent Relational Reasoning Shengyao Lu, Bang Liu, Keith G Mills, SHANGLING JUI, Di Niu

Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave Functions Nicholas Gao, Stephan Günnemann

Message Passing Neural PDE Solvers Johannes Brandstetter, Daniel E. Worrall, Max Welling

Constrained Graph Mechanics Networks Wenbing Huang, Jiaqi Han, Yu Rong, Tingyang Xu, Fuchun Sun, Junzhou Huang

Towards Training Billion Parameter Graph Neural Networks for Atomic Simulations Anuroop Sriram, Abhishek Das, Brandon M Wood, C. Lawrence Zitnick

Neural Relational Inference with Node-Specific Information Ershad Banijamali

Predicting Physics in Mesh-reduced Space with Temporal Attention XU HAN, Han Gao, Tobias Pfaff, Jian-Xun Wang, Liping Liu

## 生物&化学

Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-designWengong Jin, Jeremy Wohlwend, Regina Barzilay, Tommi S. Jaakkola

Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysis Siyi Tang, Jared Dunnmon, Khaled Kamal Saab, Xuan Zhang, Qianying Huang, Florian Dubost, Daniel Rubin, Christopher Lee-Messer

Spanning Tree-based Graph Generation for Molecules Sungsoo Ahn, Binghong Chen, Tianzhe Wang, Le Song

GeoDiff: A Geometric Diffusion Model for Molecular Conformation Generation Minkai Xu, Lantao Yu, Yang Song, Chence Shi, Stefano Ermon, Jian Tang

Independent SE(3)-Equivariant Models for End-to-End Rigid Protein Docking Octavian-Eugen Ganea, Xinyuan Huang, Charlotte Bunne, Yatao Bian, Regina Barzilay, Tommi S. Jaakkola, Andreas Krause

Spatial Graph Attention and Curiosity-driven Policy for Antiviral Drug Discovery Yulun Wu, Nicholas Choma, Andrew Deru Chen, Mikaela Cashman, Erica Teixeira Prates, Veronica G Melesse Vergara, Manesh B Shah, Austin Clyde, Thomas Brettin, Wibe Albert de Jong, Neeraj Kumar, Martha S Head, Rick L. Stevens, Peter Nugent, Daniel A Jacobson, James B Brown

Using Graph Representation Learning with Schema Encoders to Measure the Severity of Depressive Symptoms Simin Hong, Anthony Cohn, David Crossland Hogg

A Biologically Interpretable Graph Convolutional Network to Link Genetic Risk Pathways and Imaging Phenotypes of Disease Sayan Ghosal, Qiang Chen, Giulio Pergola, Aaron L Goldman, William Ulrich, Daniel R Weinberger, Archana Venkataraman

Simple GNN Regularisation for 3D Molecular Property Prediction and Beyond Jonathan Godwin, Michael Schaarschmidt, Alexander L Gaunt, Alvaro Sanchez-Gonzalez, Yulia Rubanova, Petar Veličković, James Kirkpatrick, Peter Battaglia

Geometric Transformers for Protein Interface Contact Prediction Alex Morehead, Chen Chen, Jianlin Cheng

OntoProtein: Protein Pretraining With Gene Ontology Embedding Ningyu Zhang, Zhen Bi, Xiaozhuan Liang, Siyuan Cheng, Haosen Hong, Shumin Deng, Qiang Zhang, Jiazhang Lian, Huajun Chen

Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations Keir Adams, Lagnajit Pattanaik, Connor W. Coley

Chemical-Reaction-Aware Molecule Representation Learning Hongwei Wang, Weijiang Li, Xiaomeng Jin, Kyunghyun Cho, Heng Ji, Jiawei Han, Martin Burke

Spherical Message Passing for 3D Molecular Graphs Yi Liu, Limei Wang, Meng Liu, Yuchao Lin, Xuan Zhang, Bora Oztekin, Shuiwang Ji

## 时间序列

Graph-Augmented Normalizing Flows for Anomaly Detection of Multiple Time Series Enyan Dai, Jie Chen

TAMP-S2GCNets: Coupling Time-Aware Multipersistence Knowledge Representation with Spatio-Supra Graph Convolutional Networks for Time-Series Forecasting Yuzhou Chen, Ignacio Segovia-Dominguez, Baris Coskunuzer, Yulia Gel

Space-Time Graph Neural Networks Samar Hadou, Charilaos I Kanatsoulis, Alejandro Ribeiro

Filling the G_ap_s: Multivariate Time Series Imputation by Graph Neural Networks Andrea Cini, Ivan Marisca, Cesare Alippi

Graph-Guided Network for Irregularly Sampled Multivariate Time Series Xiang Zhang, Marko Zeman, Theodoros Tsiligkaridis, Marinka Zitnik

Learning to Remember Patterns: Pattern Matching Memory Networks for Traffic Forecasting Hyunwook Lee,Seungmin Jin,Hyeshin Chu,Hongkyu Lim,Sungahn Ko

## Multi-agent

Context-Aware Sparse Deep Coordination Graphs Tonghan Wang, Liang Zeng, Weijun Dong, Qianlan Yang, Yang Yu, Chongjie Zhang

Reinforcement Learning under a Multi-agent Predictive State Representation Model: Method and Theory Zhi Zhang, Zhuoran Yang, Han Liu, Pratap Tokekar, Furong Huang

Learning Graphon Mean Field Games and Approximate Nash Equilibria Kai Cui, Heinz Koeppl

Graph-Enhanced Exploration for Goal-oriented Reinforcement Learning Jiarui Jin, Sijin Zhou, Weinan Zhang, Tong He, Yong Yu, Rasool Fakoor

Structure-Aware Transformer Policy for Inhomogeneous Multi-Task Reinforcement Learning Sunghoon Hong, Deunsol Yoon, Kee-Eung Kim

C-Planning: An Automatic Curriculum for Learning Goal-Reaching Tasks Tianjun Zhang, Benjamin Eysenbach, Ruslan Salakhutdinov, Sergey Levine, Joseph E. Gonzalez

Learning Object-Oriented Dynamics for Planning from Text Guiliang Liu, Ashutosh Adhikari, Amir-massoud Farahmand, Pascal Poupart

## Relation Modeling & Extraplation

Relational Multi-Task Learning: Modeling Relations between Data and Tasks Kaidi Cao, Jiaxuan You, Jure Leskovec

Inductive Relation Prediction Using Analogy Subgraph Embeddings Jiarui Jin, Yangkun Wang, Kounianhua Du, Weinan Zhang, Zheng Zhang, David Wipf, Yong Yu, Quan Gan

Ada-NETS: Face Clustering via Adaptive Neighbour Discovery in the Structure Space Yaohua Wang, Yaobin Zhang, Fangyi Zhang, Senzhang Wang, Ming Lin, YuQi Zhang, Xiuyu Sun

Graph-Relational Domain Adaptation Zihao Xu, Hao He, Guang-He Lee, Bernie Wang, Hao Wang

Towards Distribution Shift of Node-Level Prediction on Graphs: An Invariance Perspective Qitian Wu, Hengrui Zhang, Junchi Yan, David Wipf

MoReL: Multi-omics Relational Learning Arman Hasanzadeh, Ehsan Hajiramezanali, Nick Duffield, Xiaoning Qian

Know Your Action Set: Learning Action Relations for Reinforcement Learning Ayush Jain, Norio Kosaka, Kyung-Min Kim, Joseph J Lim

## Transformer

Revisiting Over-smoothing in BERT from the Perspective of Graph Han Shi, JIAHUI GAO, Hang Xu, Xiaodan Liang, Zhenguo Li, Lingpeng Kong, Stephen M. S. Lee, James Kwok

## Tabular Data

Convergent Boosted Smoothing for Modeling GraphData with Tabular Node Features Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Soji Adeshina, Yangkun Wang, Tom Goldstein, David Wipf

## NLP

GreaseLM: Graph REASoning Enhanced Language Models Xikun Zhang, Antoine Bosselut, Michihiro Yasunaga, Hongyu Ren, Percy Liang, Christopher D Manning, Jure Leskovec

GNN-LM: Language Modeling based on Global Contexts via GNN Yuxian Meng, Shi Zong, Xiaoya Li, Xiaofei Sun, Tianwei Zhang, Fei Wu, Jiwei Li

## Robustness

Understanding and Improving Graph Injection Attack by Promoting Unnoticeability Yongqiang Chen, Han Yang, Yonggang Zhang, MA KAILI, Tongliang Liu, Bo Han, James Cheng

Adversarial Robustness Through the Lens of Causality Yonggang Zhang, Mingming Gong, Tongliang Liu, Gang Niu, Xinmei Tian, Bo Han, Bernhard Schölkopf, Kun Zhang

## 其它

Information Gain Propagation: a New Way to Graph Active Learning with Soft Labels Wentao Zhang, Yexin Wang, Zhenbang You, Meng Cao, Ping Huang, Jiulong Shan, Zhi Yang, Bin CUI

End-to-End Learning of Probabilistic Hierarchies on Graphs Daniel Zügner, Bertrand Charpentier, Morgane Ayle, Sascha Geringer, Stephan Günnemann

Cold Brew: Distilling Graph Node Representations with Incomplete or Missing Neighborhoods Wenqing Zheng, Edward W Huang, Nikhil Rao, Sumeet Katariya, Zhangyang Wang, Karthik Subbian

Retriever: Learning Content-Style Representation as a Token-Level Bipartite Graph Dacheng Yin, Xuanchi Ren, Chong Luo, Yuwang Wang, Zhiwei Xiong, Wenjun Zeng

Neural Link Prediction with Walk Pooling Liming Pan, Cheng Shi, Ivan Dokmanić

GLASS: GNN with Labeling Tricks for Subgraph Representation Learning Xiyuan Wang, Muhan Zhang

NodePiece: Compositional and Parameter-Efficient Representations of Large Knowledge Graphs Mikhail Galkin, Etienne Denis, Jiapeng Wu, William L. Hamilton

Query Embedding on Hyper-Relational Knowledge Graphs Dimitrios Alivanistos, Max Berrendorf, Michael Cochez, Mikhail Galkin