Exponential Family Graph Signal Recovery under Structural Constraints

The problem of graph signal recovery from a relatively small number of noisy observations has been studied extensively in the recent on the field of graph signal processing. However, most of studies focus on the recovery under Gaussian noise which is just a special case of exponential family noise. It natural to ask whether we can accurately recover graph signals under different noise with exponential distribution. Herein, we extend the results of [8][9] to the more general case of exponential distribution.