Graph Convolutional Networks and Graph Attention Networks for Approximating Arguments Acceptability -- Technical Report
This work addresses computational efficiency in abstract argumentation for AI and reasoning systems, but it is incremental as it builds on the state-of-the-art AFGCN approach.
The authors tackled the problem of efficiently approximating argument acceptability in abstract argumentation by improving Graph Convolutional Networks (GCNs) for better runtime and accuracy, and further enhancing efficiency by using Graph Attention Networks (GATs).
Various approaches have been proposed for providing efficient computational approaches for abstract argumentation. Among them, neural networks have permitted to solve various decision problems, notably related to arguments (credulous or skeptical) acceptability. In this work, we push further this study in various ways. First, relying on the state-of-the-art approach AFGCN, we show how we can improve the performances of the Graph Convolutional Networks (GCNs) regarding both runtime and accuracy. Then, we show that it is possible to improve even more the efficiency of the approach by modifying the architecture of the network, using Graph Attention Networks (GATs) instead.