Unsupervised Graph Neural Architecture Search with Disentangled Self-Supervision
Authors: Zeyang Zhang, Xin Wang, Ziwei Zhang, Guangyao Shen, Shiqi Shen, Wenwu Zhu
NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on 11 real-world datasets demonstrate that the proposed DSGAS model is able to achieve state-of-the-art performance against several baseline methods in an unsupervised manner. In this section, we conduct experiments on 8 real-world datasets with unsupervised settings to verify the design of our method. |
| Researcher Affiliation | Collaboration | Zeyang Zhang1 , Xin Wang1 , Ziwei Zhang1, Guangyao Shen2, Shiqi Shen2, Wenwu Zhu1 1Department of Computer Science and Technology, BNRist, Tsinghua University, 2Wechat, Tencent |
| Pseudocode | No | The provided text of the paper does not contain any explicitly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | Yes | The codes are available at Github. |
| Open Datasets | Yes | For unsupervised settings, we conduct experiments on four graph-level classification datasets including PROTEINS [38], DD [39], MUTAG [40], IMDB-B [41] from TUDataset [42] and four node-level classification datasets Coauthor CS, Coauthor Physics from the Microsoft Academic Graph [43], Amazon Computers, Amazon Photos from the Amazon Co-purchase Graph [44]. For semi-supervised settings, we adopt three real-world datasets, OGBG-Molhiv, OGBN-Arxiv [45] and Wechat-Video. The datasets cover various graph-related fields including small molecules, bioinformatics, social networks, e-commerce networks, and academic coauthorship networks. The statistics are summarized in Table 1. |
| Dataset Splits | No | The paper states: 'Lval denotes the loss of the predictions of the architecture fα,w( ) against supervised labels on training and validation datasets.' and 'estimate the architecture performance based on the validation dataset with supervision signals.' While validation datasets are mentioned as being used, the paper does not explicitly provide specific percentages, absolute sample counts, or explicit citations to predefined splits for these datasets in the main text. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU models, CPU types, memory amounts) used for running the experiments in the main text. It mentions 'More details of the experiments are provided in the Appendix, including additional experiments and analyses, experimental setups, configurations, and implementation details.', but this statement does not explicitly describe the hardware. |
| Software Dependencies | No | The paper does not provide specific software dependency details (e.g., library names with version numbers) in the main text. It mentions 'More details of the experiments are provided in the Appendix, including additional experiments and analyses, experimental setups, configurations, and implementation details.', but this statement does not explicitly describe software versions. |
| Experiment Setup | Yes | More details of the experiments are provided in the Appendix, including additional experiments and analyses, experimental setups, configurations, and implementation details. |