Product Graph Learning from Multi-domain Data with Sparsity and Rank Constraints

In this paper, we focus on learning product graphs from multi-domain data. We assume that the graph is formed by Cartesian of two smaller graphs, which refer to as factors. pose problem estimating factor Laplacian matrices. To capture local interactions in data, seek sparse factors and a smoothness model for propose an efficient iterative solver then extend infer multi-component with applications clustering imposing rank constraints Although working computationally more attractive, not all readily admit exact factorization. end, algorithms approximate nearest graphs. The efficacy developed framework demonstrated using several numerical experiments synthetic real