Sparse Implementation of Versatile Graph-Informed Layers
This work addresses a memory bottleneck for researchers and practitioners using Graph-Informed layers in graph-structured data tasks, though it is incremental as it optimizes an existing method.
The paper tackles the inefficiency of Graph-Informed layers in graph neural networks due to dense memory allocation by introducing a sparse implementation that leverages adjacency matrix sparsity, resulting in significant memory reduction and improved computational efficiency and scalability for deeper networks and larger graphs.
Graph Neural Networks (GNNs) have emerged as effective tools for learning tasks on graph-structured data. Recently, Graph-Informed (GI) layers were introduced to address regression tasks on graph nodes, extending their applicability beyond classic GNNs. However, existing implementations of GI layers lack efficiency due to dense memory allocation. This paper presents a sparse implementation of GI layers, leveraging the sparsity of adjacency matrices to reduce memory usage significantly. Additionally, a versatile general form of GI layers is introduced, enabling their application to subsets of graph nodes. The proposed sparse implementation improves the concrete computational efficiency and scalability of the GI layers, permitting to build deeper Graph-Informed Neural Networks (GINNs) and facilitating their scalability to larger graphs.