A Novel Iterative Thresholding Algorithm for Arctangent Regularization Problem

Abstract

In this work, we derive the proximity operator of an arctangent penalty, which is expressed using hyperbolic functions of sine and cosine. This penalty is then applied to sparse signal recovery, and an efficient arctangent regularization iterative thresholding (ARIT) algorithm is proposed, offering closed-form solutions for the subproblems associated with the arctangent penalty. Extensive experiments are conducted to compare the performance of ARIT with several existing iterative thresholding algorithms, and the results demonstrate that our algorithm achieves the best overall performance in terms of the probability of successful recovery, phase transition and running time.