CCGG: A Deep Autoregressive Model for Class-Conditional Graph Generation

TL;DR

This paper introduces CCGG, a novel deep autoregressive model for class-conditional graph generation that effectively incorporates class labels to produce high-quality graphs aligned with specified features.

Contribution

The paper proposes CCGG, a new model that injects class information into graph generation and combines it with a classification loss, advancing conditional graph generation techniques.

Findings

CCGG outperforms existing methods on multiple datasets.

It maintains graph quality according to distribution-based metrics.

The approach effectively integrates class labels into graph generation.

Abstract

Graph data structures are fundamental for studying connected entities. With an increase in the number of applications where data is represented as graphs, the problem of graph generation has recently become a hot topic. However, despite its significance, conditional graph generation that creates graphs with desired features is relatively less explored in previous studies. This paper addresses the problem of class-conditional graph generation that uses class labels as generation constraints by introducing the Class Conditioned Graph Generator (CCGG). We built CCGG by injecting the class information as an additional input into a graph generator model and including a classification loss in its total loss along with a gradient passing trick. Our experiments show that CCGG outperforms existing conditional graph generation methods on various datasets. It also manages to maintain the quality of the generated graphs in terms of distribution-based evaluation metrics.