Probing Negative Sampling for Contrastive Learning to Learn Graph Representations
Abstract
Graph representation learning has long been an important yet challenging task for various real-world applications. However, its downstream tasks are mainly performed in the settings of supervised or semi-supervised learning. Inspired by recent advances in unsupervised contrastive learning, this paper is thus motivated to investigate how the node-wise contrastive learning could be performed. Particularly, we respectively resolve the class collision issue and the imbalanced negative data distribution issue. Extensive experiments are performed on three real-world datasets and the proposed approach achieves the SOTA model performance.
Cite
Text
Chen et al. "Probing Negative Sampling for Contrastive Learning to Learn Graph Representations." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021. doi:10.1007/978-3-030-86520-7_27Markdown
[Chen et al. "Probing Negative Sampling for Contrastive Learning to Learn Graph Representations." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021.](https://mlanthology.org/ecmlpkdd/2021/chen2021ecmlpkdd-probing/) doi:10.1007/978-3-030-86520-7_27BibTeX
@inproceedings{chen2021ecmlpkdd-probing,
title = {{Probing Negative Sampling for Contrastive Learning to Learn Graph Representations}},
author = {Chen, Shiyi and Wang, Ziao and Zhang, Xinni and Zhang, Xiaofeng and Peng, Dan},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2021},
pages = {434-449},
doi = {10.1007/978-3-030-86520-7_27},
url = {https://mlanthology.org/ecmlpkdd/2021/chen2021ecmlpkdd-probing/}
}