A Capsule Network-based Model for Learning Node Embeddings

TL;DR

This paper introduces Caps2NE, an unsupervised capsule network-based model that learns low-dimensional node embeddings from graph data, achieving state-of-the-art results in node classification tasks.

Contribution

The paper presents a novel capsule network architecture for node embedding that leverages a routing process to effectively aggregate neighbor features.

Findings

Caps2NE achieves state-of-the-art performance on benchmark datasets.

The capsule-based approach improves embedding quality for node classification.

Experimental results validate the effectiveness of the proposed model.

Abstract

In this paper, we focus on learning low-dimensional embeddings for nodes in graph-structured data. To achieve this, we propose Caps2NE -- a new unsupervised embedding model leveraging a network of two capsule layers. Caps2NE induces a routing process to aggregate feature vectors of context neighbors of a given target node at the first capsule layer, then feed these features into the second capsule layer to infer a plausible embedding for the target node. Experimental results show that our proposed Caps2NE obtains state-of-the-art performances on benchmark datasets for the node classification task. Our code is available at: \url{https://github.com/daiquocnguyen/Caps2NE}.