ICML2021上的图神经网络好文
发布日期：2022-03-31 04:14 点击次数：80
|
今年的ICML录取结果出了，图神经网络也是今年的一大热点，这里总结一部分我看到的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。整理不易，点个赞呗再走呗，欢迎关注我们的新专栏图神经网络实战，里面会更新一些图神经网络代码解读和最新的进展。如果大家有什么需要分享的，也欢迎投稿本专栏，感激不尽。因为好多文章还挂出来，这里先给个标题，等开会的时候再补上链接哈~GNN理论Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth Keyulu Xu (MIT) · Mozhi Zhang (University of Maryland) · Stefanie Jegelka (Massachusetts Institute of Technology) · Kenji Kawaguchi (MIT) Global optimality of linear GNN optimization w/o JK Information Obfuscation of Graph Neural Networks Peiyuan Liao (Carnegie Mellon University) · Han Zhao (University of Illinois at Urbana-Champaign) · Keyulu Xu (MIT) · Tommi Jaakkola (MIT) · Geoff Gordon (Carnegie Mellon University) · Stefanie Jegelka (Massachusetts Institute of Technology) · Ruslan Salakhutdinov (Carnegie Mellen University) fundamental tradeoffs in privacy and accuracyLet's Agree to Degree: Comparing Graph Convolutional Networks in the Message-Passing Framework Floris Geerts (University of Antwerp) · Filip Mazowiecki (MPI-SWS) · Guillermo Perez (UAntwerpen) power of GNNs with degree features and WL test?A Unified Lottery Ticket Hypothesis for Graph Neural Networks Tianlong Chen (University of Texas at Austin) · Yongduo Sui (University of Science and Technology of China) · Xuxi Chen (University of Texas at Austin) · Aston Zhang (AWS AI) · Zhangyang Wang (University of Texas at Austin)Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approach Federico Lopez (HITS - Heidelberg Institute for Theoretical Studies) · Beatrice Pozzetti (Heidelberg University) · Steve Trettel (Stanford University) · Michael Strube (Heidelberg Institute for Theoretical Studies) · Anna Wienhard (Heidelberg University)Graph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization Aseem Baranwal (University of Waterloo) · Kimon Fountoulakis (University of Waterloo) · Aukosh Jagannath (University of Waterloo)Breaking the Limits of Message Passing Graph Neural Networks Muhammet Balcilar (Université de Rouen - LITIS) · Pierre Heroux (University of Rouen Normandy) · Benoit Gauzere (INSA Rouen) · Sebastien Adam (Université de Rouen Normandie) · Paul Honeine (LITIS Lab, Université de Rouen Normandie) · Pascal Vasseur (LITIS Université de Rouen Normandie) looks very exciting; not available yetFrom Local Structures to Size Generalization in Graph Neural Networks Gilad Yehudai (Weizmann Institute of Science) · Ethan Fetaya (Bar-Ilan University) · eli meirom (NVIDIA) · Gal Chechik (Nvidia) · Haggai Maron (NVIDIA Research) size generalization and d-patternGNNAutoScale: Scalable and Expressive Graph Neural Networks via Historical Embeddings Matthias Fey (TU Dortmund University) · Jan Eric Lenssen (TU Dortmund) · Frank Weichert (Technical University of Dortmund) · Jure Leskovec (Stanford University) seems size invariant and preserving expressive power by reading abstractDirected Graph Embeddings in Pseudo-Riemannian Manifolds Aaron Sim (BenevolentAI) · Maciej Wiatrak (BenevolentAI) · Angus Brayne (BenevolentAI) · Páidí Creed (BenevolentAI) · Saee Paliwal (Benevolent AI)Interpretable Stability Bounds for Spectral Graph Filters Henry Kenlay (University of Oxford) · Dorina Thanou (Swiss Data Science Center (EPFL and ETH Zurich)) · Xiaowen Dong (University of Oxford)DeepWalking Backwards: From Embeddings Back to Graphs Sudhanshu Chanpuriya (University of Massachusetts Amherst) · Cameron Musco (University of Massachusetts Amherst) · Konstantinos Sotiropoulos (Boston University) · Charalampos Tsourakakis (ISI Foundation, Boston University) Embedding as matrix factorization and an exact factorization algorithm新的GNN结构GRAND: Graph Neural Diffusion Ben Chamberlain (Twitter) · Maria Gorinova (University of Edinburgh) · Michael Bronstein (Twitter) · Stefan Webb (Twitter) · James Rowbottom (Twitter) · Emanuele Rossi (Twitter)How Framelets Enhance Graph Neural Networks Xuebin Zheng (The University of Sydney) · Bingxin Zhou (The University of Sydney) · Junbin Gao (The University of Sydney) · Yuguang Wang (Max Planck Institute for Mathematics in Sciences; Shanghai Jiao Tong University; University of New South Wales) · Pietro Lió (University of Cambridge) · Ming Li (Zhejiang Normal University) · Guido Montufar (UCLA Math / Stat; MPI MIS)Directional Graph Networks Dominique Beaini (InVivo AI) · Saro Passaro (University of Cambridge) · Vincent Létourneau (Université de Ottawa) · Will Hamilton (McGill University and Mila) · Gabriele Corso (University of Cambridge) · Pietro Lió (University of Cambridge)Training Graph Neural Networks with 1000 Layers Guohao Li (KAUST) · Matthias Müller (Intel Labs) · Bernard Ghanem (KAUST) · Vladlen Koltun (Intel Labs) exciting results; reversible connections, group convolutions, weight tying, and equilibrium models to advance the memory and parameter efficiencyGraph Mixture Density NetworksFederico Errica (University of Pisa) · Davide Bacciu (University of Pisa) · Alessio Micheli (Universita di Pisa)E(n) Equivariant Graph Neural Networks Víctor Garcia Satorras (University of Amsterdam) · Emiel Hoogeboom (University of Amsterdam) · Max Welling (University of Amsterdam & Qualcomm)Lipschitz normalization for self-attention layers with application to graph neural networks George Dasoulas (Ecole Polytechnique, Paris, France) · Kevin Scaman (Noah's Ark, Huawei Technologies) · Aladin Virmaux (Huawei)GNNAutoScale: Scalable and Expressive Graph Neural Networks via Historical Embeddings Matthias Fey (TU Dortmund University) · Jan Eric Lenssen (TU Dortmund) · Frank Weichert (Technical University of Dortmund) · Jure Leskovec (Stanford University)Elastic Graph Neural Networks Xiaorui Liu (Michigan State University) · Wei Jin (Michigan State University) · Yao Ma (Michigan State University) · Yaxin Li (Michigan State University) · Hua Liu (Shandong University ) · Yiqi Wang (Michigan State University) · Ming Yan (Michigan State University) · Jiliang Tang (Michigan State University)Size-Invariant Graph Representations for Graph Classification Extrapolations Beatrice Bevilacqua (Purdue University) · Yangze Zhou (Purdue University) · Bruno Ribeiro (Purdue University)Graph Neural Networks Inspired by Classical Iterative Algorithms Yang Yongyi (Fudan University) · Tang Liu (Fudan University) · Yangkun Wang (SJTU) · Jinjing Zhou (Amazon) · Quan Gan (Amazon) · Zhewei Wei (Renmin University of China) · Zheng Zhang (Amazon) · Zengfeng Huang (Fudan University) · David Wipf (Microsoft Research) unrolled proximal algorithms as GNNsGNN与组合优化Deep Latent Graph Matching Tianshu Yu (Arizona State University) · Runzhong Wang (Shanghai Jiao Tong University) · Junchi Yan (Shanghai Jiao Tong University) · baoxin Li (Arizona State University)On Explainability of Graph Neural Networks via Subgraph Explorations Hao Yuan (Texas A&M University) · Haiyang Yu (Texas A&M University) · Jie Wang (University of Science and Technology of China) · Kang Li (Rutgers) · Shuiwang Ji (Texas A&M University) explanation as combinatorial optimization over graph如何更好地训练GNNGraphNorm: A Principled Approach to Accelerating Graph Neural Network Training Tianle Cai (Princeton University) · Shengjie Luo (Peking University) · Keyulu Xu (MIT) · Di He (Microsoft Research) · Tie-Yan Liu (Microsoft Research Asia) · Liwei Wang (Peking University)Self-supervised Graph-level Representation Learning with Local and Global Structure Minghao Xu (Shanghai Jiao Tong University) · Hang Wang (Shanghai Jiao Tong University) · Bingbing Ni (Shanghai Jiao Tong University) · Hongyu Guo (National Research Council Canada) · Jian Tang (HEC Montreal & MILA)Memory-Efficient Graph Neural Networks Guohao Li (KAUST) · Matthias Müller (Intel Labs) · Bernard Ghanem (KAUST) · Vladlen Koltun (Intel Labs) exciting results; idea seems like neural ode by only reading abstract?Graph Contrastive Learning Automated Yuning You (Texas A&M University) · Tianlong Chen (University of Texas at Austin) · Yang Shen (Texas A&M University) · Zhangyang Wang (University of Texas at Austin)From Local Structures to Size Generalization in Graph Neural Networks Gilad Yehudai (Weizmann Institute of Science) · Ethan Fetaya (Bar-Ilan University) · eli meirom (NVIDIA) · Gal Chechik (Nvidia) · Haggai Maron (NVIDIA Research) new self-supervised taskImproving Breadth-Wise Backpropagation in Graph Neural Networks helps Learning Long-Range Dependencies. Denis Lukovnikov (University of Bonn) · Asja Fischer (Ruhr University Bochum)Automated Graph Representation Learning with Hyperparameter Importance Explanation Xin Wang (Tsinghua University) · Shuyi Fan (Tsinghua University) · Kun Kuang (Tsinghua University) · wenwu zhu (Tsinghua University)GNN ExtrapolationGraph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization Aseem Baranwal (University of Waterloo) · Kimon Fountoulakis (University of Waterloo) · Aukosh Jagannath (University of Waterloo)From Local Structures to Size Generalization in Graph Neural Networks Gilad Yehudai (Weizmann Institute of Science) · Ethan Fetaya (Bar-Ilan University) · eli meirom (NVIDIA) · Gal Chechik (Nvidia) · Haggai Maron (NVIDIA Research) new self-supervised taskGNNAutoScale: Scalable and Expressive Graph Neural Networks via Historical Embeddings Matthias Fey (TU Dortmund University) · Jan Eric Lenssen (TU Dortmund) · Frank Weichert (Technical University of Dortmund) · Jure Leskovec (Stanford University)Size-Invariant Graph Representations for Graph Classification Extrapolations Beatrice Bevilacqua (Purdue University) · Yangze Zhou (Purdue University) · Bruno Ribeiro (Purdue University)GNN可解释性On Explainability of Graph Neural Networks via Subgraph Explorations Hao Yuan (Texas A&M University) · Haiyang Yu (Texas A&M University) · Jie Wang (University of Science and Technology of China) · Kang Li (Rutgers) · Shuiwang Ji (Texas A&M University) explanation as combinatorial optimization over graphGenerative Causal Explanations for Graph Neural Networks Wanyu LIN (University of Toronto) · Hao Lan (University of Toronto) · Baochun Li (University of Toronto) explanation as graph generationImproving Molecular Graph Neural Network Explainability with Orthonormalization and Induced Sparsity Ryan Henderson (Bayer) · Djork-Arné Clevert (Bayer AG) · Floriane Montanari (Bayer AG)图生成GraphDF: A Discrete Flow Model for Molecular Graph Generation Youzhi Luo (Texas A&M University) · Keqiang Yan (Texas A&M University, College Station) · Shuiwang Ji (Texas A&M University)Order Matters: Probabilistic Modeling of Node Sequence for Graph Generation Xiaohui Chen (Tufts University) · Xu Han (Tufts University) · Jiajing Hu (Tufts University) · Francisco R Ruiz (DeepMind) · Liping Liu (Tufts University)对抗攻击Expressive 1-Lipschitz Neural Networks for Robust Multiple Graph Learning against Adversarial Attacks Xin Zhao (Auburn University) · Zeru Zhang (Auburn University) · Zijie Zhang (Auburn University) · Lingfei Wu (IBM Research AI) · Jiayin Jin (Auburn University) · Yang Zhou (Auburn University) · Ruoming Jin (Kent State University) · Dejing Dou (" University of Oregon, USA") · Da Yan (University of Alabama at Birmingham)Information Obfuscation of Graph Neural Networks Peiyuan Liao (Carnegie Mellon University) · Han Zhao (University of Illinois at Urbana-Champaign) · Keyulu Xu (MIT) · Tommi Jaakkola (MIT) · Geoff Gordon (Carnegie Mellon University) · Stefanie Jegelka (Massachusetts Institute of Technology) · Ruslan Salakhutdinov (Carnegie Mellen University) fundamental tradeoffs in robustness and accuracyIntegrated Defense for Resilient Graph Matching Jiaxiang Ren (Auburn University) · Zijie Zhang (Auburn University) · Jiayin Jin (Auburn University) · Xin Zhao (Auburn University) · Sixing Wu (Peking University) · Yang Zhou (Auburn University) · Yelong Shen (Microsoft Dynamics 365 AI) · Tianshi Che (Auburn University) · Ruoming Jin (Kent State University) · Dejing Dou (" University of Oregon, USA")GNN与物理Controlling Graph Dynamics with Reinforcement Learning and Graph Neural Networks Eli Meirom (NVIDIA Research) · Haggai Maron (NVIDIA Research) · Shie Mannor (Technion) · Gal Chechik (NVIDIA / Bar-Ilan University)GNN与时间序列Z-GCNETs: Time Zigzags at Graph Convolutional Networks for Time Series Forecasting Yuzhou Chen (Southern Methodist University) · Ignacio Segovia (University of Texas at Dallas) · Yulia R Gel (University of Texas at Dallas)GNN与auto-MLAutomated Graph Representation Learning with Hyperparameter Importance Explanation Xin Wang (Tsinghua University) · Shuyi Fan (Tsinghua University) · Kun Kuang (Tsinghua University) · wenwu zhu (Tsinghua University)看标题没看懂咋分类的World Model as a Graph: Learning Latent Landmarks for Planning Lunjun Zhang (University of Toronto) · Ge Yang (University of Chicago) · Bradly Stadie (Vector Institute)
现如今打麻将技巧，喜欢麻将游戏的玩家是越来越多了，但是抱怨不可以赢得麻将游戏的玩家也是逐渐的在增多。其实玩家们会觉得麻将游戏很难赢的唯一原因就是他们都不知道有打麻将赢钱秘诀...
|
