Hi-PART: Going Beyond Graph Pooling with Hierarchical Partition Tree for Graph-Level Representation Learning
文献类型：期刊论文
| 作者 | Ren, Yuyang1; Zhang, Haonan1; Fu, Luoyi1; Liang, Shiyu1; Zhou, Lei1; Wang, Xinbing1; Cao, Xinde1; Long, Fei2; Zhou, Chenghu3 |
| 刊名 | ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
 |
| 出版日期 | 2024-05-01 |
| 卷号 | 18期号:4页码:20 |
| 关键词 | Graph representation learning structural entropy |
| ISSN号 | 1556-4681 |
| DOI | 10.1145/3636429 |
| 通讯作者 | Ren, Yuyang(renyuyang@sjtu.edu.cn) |
| 英文摘要 | Graph pooling refers to the operation that maps a set of node representations into a compact form for graph-level representation learning. However, existing graph pooling methods are limited by the power of the Weisfeiler-Lehman (WL) test in the performance of graph discrimination. In addition, these methods often suffer from hard adaptability to hyper-parameters and training instability. To address these issues, we propose Hi-PART, a simple yet effective graph neural network (GNN) framework with Hierarchical Partition Tree (HPT). In HPT, each layer is a partition of the graph with different levels of granularities that are going toward a finer grain from top to bottom. Such an exquisite structure allows us to quantify the graph structure information contained in HPT with the aid of structural information theory. Algorithmically, by employing GNNs to summarize node features into the graph feature based on HPT's hierarchical structure, Hi-PART is able to adequately leverage the graph structure information and provably goes beyond the power of the WL test. Due to the separation of HPT optimization from graph representation learning, Hi-PART involves the height of HPT as the only extra hyper-parameter and enjoys higher training stability. Empirical results on graph classification benchmarks validate the superior expressive power and generalization ability of Hi-PART compared with state-of-the-art graph pooling approaches. |
| 资助项目 | NSF China[62020106005] ; NSF China[61960206002] ; NSF China[42050105] ; NSF China[62061146002] ; Shanghai Pilot Program for Basic Research - Shanghai Jiao Tong University |
| WOS研究方向 | Computer Science |
| 语种 | 英语 |
| WOS记录号 | WOS:001190988100022 |
| 出版者 | ASSOC COMPUTING MACHINERY |
| 资助机构 | NSF China ; Shanghai Pilot Program for Basic Research - Shanghai Jiao Tong University |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/204218]  |
| 专题 | 中国科学院地理科学与资源研究所 |
| 通讯作者 | Ren, Yuyang |
| 作者单位 | 1.Shanghai Jiao Tong Univ, 800 Dong chuan Rd, Shanghai 200240, Peoples R China 2.Xinhua News Agcy, State Key Lab Media Convergence Prod Technol & Sy, 9 Beizhenghongqi West St, Beijing 100803, Peoples R China 3.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, 11 Datun Rd, Beijing 100101, Peoples R China |
| 推荐引用方式 GB/T 7714 | Ren, Yuyang,Zhang, Haonan,Fu, Luoyi,et al. Hi-PART: Going Beyond Graph Pooling with Hierarchical Partition Tree for Graph-Level Representation Learning[J]. ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA,2024,18(4):20. |
| APA | Ren, Yuyang.,Zhang, Haonan.,Fu, Luoyi.,Liang, Shiyu.,Zhou, Lei.,...&Zhou, Chenghu.(2024).Hi-PART: Going Beyond Graph Pooling with Hierarchical Partition Tree for Graph-Level Representation Learning.ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA,18(4),20. |
| MLA | Ren, Yuyang,et al."Hi-PART: Going Beyond Graph Pooling with Hierarchical Partition Tree for Graph-Level Representation Learning".ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA 18.4(2024):20. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。