The construction of a meaningful graph plays a crucial role in the success of many graph-based data representations and algorithms, especially in the emerging field of signal processing on graphs. However, a meaningful graph is not always readily available from the data, nor easy to define depending on the application domain. In this paper, we address the problem of graph learning, where we are interested in learning graph topologies, namely, the relationships between data entities, that well explain the signal observations. In particular, we want to infer a graph such that the input data forms graph signals with smooth variations on the resulting topology. To this end, we adopt a factor analysis model for the graph signals and impose a Gaussian probabilistic prior on the latent variables that control these graph signals. We show that the Gaussian prior leads to an efficient representation that favors the smoothness property of the graph signals. We then propose an algorithm for learning graphs that enforce such smoothness property for the signal observations by minimizing the variations of the signals on the learned graph. Experiments on both synthetic and real world data demonstrate that the proposed graph learning framework can efficiently infer meaningful graph topologies from only the signal observations.
View on arXiv