Heterogeneous Graph Neural Networks for Assumption-Based Argumentation

TL;DR

This paper introduces novel GNN architectures, ABAGCN and ABAGAT, to efficiently approximate credulous acceptance in Assumption-Based Argumentation, outperforming previous models and enabling scalable reasoning in large frameworks.

Contribution

First GNN-based approach for approximating credulous acceptance in ABA, with new architectures and a polynomial-time extension-reconstruction algorithm.

Findings

ABAGCN and ABAGAT outperform baseline GNN models.

Achieve up to 0.71 node-level F1 score on ICCMA instances.

Reconstruct stable extensions with F1 above 0.85 on small ABAFs.

Abstract

Assumption-Based Argumentation (ABA) is a powerful structured argumentation formalism, but exact computation of extensions under stable semantics is intractable for large frameworks. We present the first Graph Neural Network (GNN) approach to approximate credulous acceptance in ABA. To leverage GNNs, we model ABA frameworks via a dependency graph representation encoding assumptions, claims and rules as nodes, with heterogeneous edge labels distinguishing support, derive and attack relations. We propose two GNN architectures - ABAGCN and ABAGAT - that stack residual heterogeneous convolution or attention layers, respectively, to learn node embeddings. Our models are trained on the ICCMA 2023 benchmark, augmented with synthetic ABAFs, with hyperparameters optimised via Bayesian search. Empirically, both ABAGCN and ABAGAT outperform a state-of-the-art GNN baseline that we adapt from the abstract argumentation literature, achieving a node-level F1 score of up to 0.71 on the ICCMA instances. Finally, we develop a sound polynomial time extension-reconstruction algorithm driven by our predictor: it reconstructs stable extensions with F1 above 0.85 on small ABAFs and maintains an F1 of about 0.58 on large frameworks. Our work opens new avenues for scalable approximate reasoning in structured argumentation.