Empirically Accelerating Scaled Gradient Projection Using Deep Neural Network For Inverse Problems In Image Processing
This work addresses the need for faster and more reliable optimization in image processing, though it appears incremental as it builds on existing scaled gradient projection methods.
The paper tackles the problem of accelerating optimization algorithms for large-scale inverse problems in image processing by training a deep neural network to yield parameters in the scaled gradient projection method, which were previously chosen heuristically. The result is a novel DNN-based convergent iterative algorithm that significantly improves the empirical convergence rate over conventional methods.
Recently, deep neural networks (DNNs) have shown advantages in accelerating optimization algorithms. One approach is to unfold finite number of iterations of conventional optimization algorithms and to learn parameters in the algorithms. However, these are forward methods and are indeed neither iterative nor convergent. Here, we present a novel DNN-based convergent iterative algorithm that accelerates conventional optimization algorithms. We train a DNN to yield parameters in scaled gradient projection method. So far, these parameters have been chosen heuristically, but have shown to be crucial for good empirical performance. In simulation results, the proposed method significantly improves the empirical convergence rate over conventional optimization methods for various large-scale inverse problems in image processing.