Revisiting Semi-Supervised Learning with Graph Embeddings

TL;DR

This paper introduces a semi-supervised learning framework using graph embeddings that effectively predicts labels for both seen and unseen instances across various tasks, improving upon existing models.

Contribution

It develops both transductive and inductive graph embedding methods for semi-supervised learning, enabling predictions on new data and demonstrating superior performance on multiple benchmarks.

Findings

Improved accuracy over existing models on diverse tasks

Effective handling of unseen instances with inductive embeddings

Versatile framework applicable to text and entity classification

Abstract

We present a semi-supervised learning framework based on graph embeddings. Given a graph between instances, we train an embedding for each instance to jointly predict the class label and the neighborhood context in the graph. We develop both transductive and inductive variants of our method. In the transductive variant of our method, the class labels are determined by both the learned embeddings and input feature vectors, while in the inductive variant, the embeddings are defined as a parametric function of the feature vectors, so predictions can be made on instances not seen during training. On a large and diverse set of benchmark tasks, including text classification, distantly supervised entity extraction, and entity classification, we show improved performance over many of the existing models.