OLGA: One-cLass Graph Autoencoder

TL;DR

OLGA is an end-to-end graph autoencoder for one-class learning that improves node representation, encapsulates interest instances with a novel hypersphere loss, and offers interpretable low-dimensional embeddings, outperforming existing methods.

Contribution

OLGA introduces a new hypersphere loss for one-class graph learning, combining it with autoencoder reconstruction for improved, interpretable node representations.

Findings

Achieved state-of-the-art results on benchmark datasets.

Outperformed six other methods with statistical significance.

Produced low-dimensional, interpretable representations without sacrificing performance.

Abstract

One-class learning (OCL) comprises a set of techniques applied when real-world problems have a single class of interest. The usual procedure for OCL is learning a hypersphere that comprises instances of this class and, ideally, repels unseen instances from any other classes. Besides, several OCL algorithms for graphs have been proposed since graph representation learning has succeeded in various fields. These methods may use a two-step strategy, initially representing the graph and, in a second step, classifying its nodes. On the other hand, end-to-end methods learn the node representations while classifying the nodes in one learning process. We highlight three main gaps in the literature on OCL for graphs: (i) non-customized representations for OCL; (ii) the lack of constraints on hypersphere parameters learning; and (iii) the methods' lack of interpretability and visualization. We propose One-cLass Graph Autoencoder (OLGA). OLGA is end-to-end and learns the representations for the graph nodes while encapsulating the interest instances by combining two loss functions. We propose a new hypersphere loss function to encapsulate the interest instances. OLGA combines this new hypersphere loss with the graph autoencoder reconstruction loss to improve model learning. OLGA achieved state-of-the-art results and outperformed six other methods with a statistically significant difference from five methods. Moreover, OLGA learns low-dimensional representations maintaining the classification performance with an interpretable model representation learning and results.