Abstract
Graph-based semi-supervised learning is appealing when labels are scarce but large amounts of unlabeled data are available. These methods typically use a heuristic strategy to construct the graph based on some fixed data representation, independently of the available labels. In this pa- per, we propose to jointly learn a data representation and a graph from both labeled and unlabeled data such that (i) the learned representation indirectly encodes the label information injected into the graph, and (ii) the graph provides a smooth topology with respect to the transformed data. Plugging the resulting graph and representation into existing graph-based semi-supervised learn- ing algorithms like label spreading and graph convolutional networks, we show that our approach outperforms standard graph construction methods on both synthetic data and real datasets.- Anthology ID:
- 2020.textgraphs-1.4
- Volume:
- Proceedings of the Graph-based Methods for Natural Language Processing (TextGraphs)
- Month:
- December
- Year:
- 2020
- Address:
- Barcelona, Spain (Online)
- Venue:
- TextGraphs
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 35–45
- Language:
- URL:
- https://aclanthology.org/2020.textgraphs-1.4
- DOI:
- 10.18653/v1/2020.textgraphs-1.4
- Cite (ACL):
- Mariana Vargas-Vieyra, Aurélien Bellet, and Pascal Denis. 2020. Joint Learning of the Graph and the Data Representation for Graph-Based Semi-Supervised Learning. In Proceedings of the Graph-based Methods for Natural Language Processing (TextGraphs), pages 35–45, Barcelona, Spain (Online). Association for Computational Linguistics.
- Cite (Informal):
- Joint Learning of the Graph and the Data Representation for Graph-Based Semi-Supervised Learning (Vargas-Vieyra et al., TextGraphs 2020)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2020.textgraphs-1.4.pdf