Graph Neural Networks with Soft Association between Topology and Attribute
Authors: Yachao Yang, Yanfeng Sun, Shaofan Wang, Jipeng Guo, Junbin Gao, Fujiao Ju, Baocai Yin
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on homophilic and heterophilic graph datasets convincingly demonstrate that the proposed GNN-SATA effectively captures more accurate adjacency relationships and outperforms state-of-the-art approaches. |
| Researcher Affiliation | Academia | 1Beijing Key Laboratory of Multimedia and Intelligent Software Technology, Beijing Institute of Artificial Intelligence, Faculty of Information Technology, Beijing University of Technology, Beijing, China 2College of Information Science and Technology, Beijing University of Chemical Technology, China 3Discipline of Business Analytics, The University of Sydney Business School, The University of Sydney, Camperdown, NSW 2006, Australia. |
| Pseudocode | Yes | Algorithm 1: Optimization Algorithm of GNN-SATA |
| Open Source Code | Yes | Our code is released at https://github.com/wwwfadecom/GNN-SATA. |
| Open Datasets | Yes | The proposed GNN-SATA is evaluated on four high homophilic datasets (Cora, Citeseer, Photo, Computer) and two heterophilic (Squirrel, Chameleon) datasets. The attribute statistics of datasets are shown in Table 1. |
| Dataset Splits | Yes | In order to assess the effectiveness of the proposed model in classification tasks, we partitioned the nodes of each class in all datasets into three sets: 60% for training, 20% for validation, and 20% for testing like other baselines. |
| Hardware Specification | Yes | The training process iterates for 200 epochs on a machine with RTX 3090 Ti GPU. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., Python version, library versions). |
| Experiment Setup | Yes | The training process iterates for 200 epochs on a machine with RTX 3090 Ti GPU. ... The value of parameter α1 is chosen from the set [1, 3, 5, 7, 10], while parameter α2 is selected from the set [0.001, 0.01, 0.1, 1]. |