Graph Oriented Attention Networks

Abstract

Graph Attention Networks (GAT) is a type of neural network architecture designed to effectively model and process data represented as graphs. GATs leverage the concept of attention mechanisms to learn the importance of different nodes in a graph when performing tasks such as node classification or link prediction. By assigning attention weights to neighboring nodes, GATs can capture the relevance and influence of different graph elements in a localized manner. In this paper, we present Graph Oriented Attention Networks (GOAT) as a novel approach for attention calculation in graph-related tasks, where attention is specifically directed towards a particular destination node during each iteration. This approach aims to enhance the interpretability and control of attention within a graph neural network. By focusing attention on a specific destination node, the model can explicitly consider the influence of that particular node on the different graph nodes, providing insights into its role and importance within the graph structure. This approach is particularly valuable in scenarios where understanding the impact of individual nodes on specific targets is critical: in our case, to predict a heuristic for the $\text{A}^{\ast} $ search algorithm, we are focusing the attention on the destination node. Experimental results indicate that this new approach effectively improves both the performance and interpretability of graph-based models, facilitating a more fine-grained analysis of graph dynamics and the factors influencing specific nodes.

Keywords

• Attention mechanism
• graph neural networks
• graph theory
• heuristics
• Python