In this paper, a graph-theory-based approach for representing planar mechanisms is presented, the Santiago Portilla method (SPM). From the corresponding adjacency matrix, SPM generates an extended ...

This paper introduces a novel graph neural layer, the dynamic connection layer (DCL), designed to address chemical graphs’ inaccurate atomic connection. The DCL layer employs a correction function, ...

Contrastive learning has been widely used in graph representation learning, which extracts node or graph representations by contrasting positive and negative node pairs. It requires node ...