Graph Vertex Embeddings: Distance, Regularization and Community Detection
This work addresses the challenge of efficiently representing graph structures for tasks like community detection, offering incremental improvements over existing methods.
The paper tackled the problem of improving graph vertex embeddings by introducing flexible distance functions and analyzing embeddings as transformations of distance matrices, resulting in competitive community detection performance on benchmark datasets with reduced computational complexity.
Graph embeddings have emerged as a powerful tool for representing complex network structures in a low-dimensional space, enabling the use of efficient methods that employ the metric structure in the embedding space as a proxy for the topological structure of the data. In this paper, we explore several aspects that affect the quality of a vertex embedding of graph-structured data. To this effect, we first present a family of flexible distance functions that faithfully capture the topological distance between different vertices. Secondly, we analyze vertex embeddings as resulting from a fitted transformation of the distance matrix rather than as a direct result of optimization. Finally, we evaluate the effectiveness of our proposed embedding constructions by performing community detection on a host of benchmark datasets. The reported results are competitive with classical algorithms that operate on the entire graph while benefitting from a substantially reduced computational complexity due to the reduced dimensionality of the representations.