When Does Self-Supervision Help Graph Convolutional Networks?
Authors: Yuning You, Tianlong Chen, Zhangyang Wang, Yang Shen
ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our results show that, with properly designed task forms and incorporation mechanisms, self-supervision benefits GCNs in gaining more generalizability and robustness. Our codes are available at https: //github.com/Shen-Lab/SS-GCNs. 4. Experiments |
| Researcher Affiliation | Academia | 1Texas A&M University. Correspondence to: Yang Shen <yshen@tamu.edu>. |
| Pseudocode | No | No structured pseudocode or algorithm blocks were found. |
| Open Source Code | Yes | Our codes are available at https: //github.com/Shen-Lab/SS-GCNs. |
| Open Datasets | Yes | Cora, Citeseer, Pub Med |
| Dataset Splits | Yes | In this paper, we mainly focus our discussion on transductive semisupervised node classification, as a representative testbed for GCNs, where there are abundant unlabeled nodes and a small number of labeled nodes in the graph, with the target to predict the labels of remaining unlabeled nodes. (Kipf & Welling, 2016) on the aspects of: 1) the standard performances of GCN (Kipf & Welling, 2016) with different self-supervision schemes; ... Table 2: Experiments for GCN through M3S. Gray numbers are from (Sun et al., 2019). Label Rate 0.03% 0.1% 0.3% (Conventional dataset split) ... Table 5: Node classification performances (accuracy; unit: %) when incorporating three self-supervision tasks (Node Clustering, Graph Partitioning, and Graph Completion) into GCNs through various schemes: pretraining & finetuning (abbr. P&T), selftraining M3S (Sun et al., 2019)), and multi-task learning (abbr. MTL). Red numbers indicate the best two performances with the mean improvement at least 0.8 (where 0.8 is comparable or less than observed standard deviations). In the case of GCN without self-supervision, gray numbers indicate the published results. ... Each combination of self-supervised scheme and task is run 50 times for each dataset with different random seeds so that the mean and the standard deviation of its performance can be reported. |
| Hardware Specification | No | No specific hardware details (e.g., GPU/CPU models, memory) were provided for running the experiments. |
| Software Dependencies | No | No specific software dependencies with version numbers were mentioned. |
| Experiment Setup | Yes | Implementation details can be found in Appendix A. |