1: \begin{abstract}

2: This work aims to accelerate the convergence of proximal gradient methods used

3: to solve regularized linear inverse problems.

4: This is achieved by designing a polynomial-based preconditioner that targets

5: the eigenvalue spectrum of the normal operator derived from the linear

6: operator.

7: The preconditioner does not assume any explicit structure on the linear

8: function and thus can be deployed in diverse applications of interest.

9: The efficacy of the preconditioner is validated on three different Magnetic

10: Resonance Imaging applications, where it is seen to achieve faster iterative

11: convergence while achieving similar reconstruction quality.

12: \end{abstract}

13: