Effective Stabilized Self-Training on Few-Labeled Graph Data Article Swipe

MAKE IMPACT

Ziang Zhou , Shenzhong Zhang , Zengfeng Huang · YOU? · · 2019 · Open Access · · DOI: https://doi.org/10.48550/arxiv.1910.02684

Graph neural networks (GNNs) are designed for semi-supervised node classification on graphs where only a subset of nodes have class labels. However, under extreme cases when very few labels are available (e.g., 1 labeled node per class), GNNs suffer from severe performance degradation. Specifically, we observe that existing GNNs suffer from unstable training process on few-labeled graphs, resulting to inferior performance on node classification. Therefore, we propose an effective framework, Stabilized Self-Training (SST), which is applicable to existing GNNs to handle the scarcity of labeled data, and consequently, boost classification accuracy. We conduct thorough empirical and theoretical analysis to support our findings and motivate the algorithmic designs in SST. We apply SST to two popular GNN models GCN and DAGNN, to get SSTGCN and SSTDA methods respectively, and evaluate the two methods against 10 competitors over 5 benchmarking datasets. Extensive experiments show that the proposed SST framework is highly effective, especially when few labeled data are available. Our methods achieve superior performance under almost all settings over all datasets. For instance, on a Cora dataset with only 1 labeled node per class, the accuracy of SSTGCN is 62.5%, 17.9% higher than GCN, and the accuracy of SSTDA is 66.4%, which outperforms DAGNN by 6.6%.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/1910.02684

PDF

https://arxiv.org/pdf/1910.02684

OA Status

green

Cited By

6

References

62

Related Works

10

OpenAlex ID

https://openalex.org/W2978310188

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W2978310188

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.1910.02684

Digital Object Identifier

Title

Effective Stabilized Self-Training on Few-Labeled Graph Data

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2019

Year of publication

Publication date

2019-10-07

Full publication date if available

Authors

Ziang Zhou, Shenzhong Zhang, Zengfeng Huang

List of authors in order

Landing page

https://arxiv.org/abs/1910.02684

Publisher landing page

PDF URL

https://arxiv.org/pdf/1910.02684

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/1910.02684

Direct OA link when available

Concepts

Computer science, Benchmarking, Graph, Labeled data, Node (physics), Artificial intelligence, Machine learning, Data mining, Class (philosophy), Training set, Theoretical computer science, Engineering, Marketing, Business, Structural engineering

Top concepts (fields/topics) attached by OpenAlex

Cited by

6

Total citation count in OpenAlex

Citations by year (recent)

2024: 1, 2023: 1, 2021: 2, 2020: 1, 2019: 1

Per-year citation counts (last 5 years)

References (count)

62

Number of works referenced by this work

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload