OPEN: Orthogonal Propagation with Ego-Network Modeling
Authors: Liang Yang, Lina Kang, Qiuliang Zhang, Mengzhe Li, bingxin niu, Dongxiao He, Zhen Wang, Chuan Wang, Xiaochun Cao, Yuanfang Guo
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 4 Evaluations Firstly, this section provides experimental setups, including dataset, baselines and implementation details. Then, the node classification results are analyzed followed by hyper-parameter tuning. Finally, the capabilities on preventing over-smoothing and overfitting are verified. |
| Researcher Affiliation | Academia | 1School of Artificial Intelligence, Hebei University of Technology, Tianjin, China 2College of Intelligence and Computing, Tianjin University, Tianjin, China 3 School of Artificial Intelligence, OPtics and Electro Nics (i OPEN), Northwestern Polytechnical University, Xi an, China 4School of Cybersecurity, Northwestern Polytechnical University, Xi an, China 5State Key Laboratory of Information Security, IIE, CAS, Beijing, China 6School of Cyber Science and Technology, Sun Yat-sen University, Shenzhen, China 7School of Computer Science and Engineering, Beihang University, Beijing, China 8Zhongguancun Laboratory, Beijing, China |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | 3. If you ran experiments... (a) Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Yes] |
| Open Datasets | Yes | To comprehensively evaluate the proposed OPEN, 7 widely used datasets are employed. Statistics of datasets are shown in Table 1. These datasets can be divided into three categories. Citation Networks. Cora, Citeseer, and Pubmed, which are widely used to verify GNNs, are standard citation network benchmark datasets [34, 35]. Co-purchase Networks. Amazon-Computers (Computers) and Amazon-Photo (Photo) are two networks of co-purchase relationships [36]. Coauthor Networks. Coauthor-CS (CS) and Coauthor-Physics (Physics) are two co-author networks based on the Microsoft Academic Graph from the KDD Cup 2016 challenge [36]. |
| Dataset Splits | Yes | For all datasets, we randomly split nodes of each class in to 60%, 20% and 20% for training validation and testing, and run on test sets over 10 random splits, as suggested in [42]. |
| Hardware Specification | No | The paper does not specify any hardware details such as GPU or CPU models, or memory used for the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers. |
| Experiment Setup | Yes | The proposed OPEN employs 2-layers network with K = 5 channels except for the hyper-parameter tuning (Sec. 4.5) and over-smoothing investigation (Sct 4.6). The whole network is trained in an end-to-end manner using the Adam optimizer with an initial learning rate of 0.001. The maximum number of epochs is set up to 1000. Besides, early stopping with a patience of 50 is also utilized. |