Convergence analysis of equilibrium methods for inverse problems
This provides a theoretical foundation for modern regularization techniques in fields like medical imaging and remote sensing, though it is incremental as it extends existing variational theory.
The paper tackles the theoretical gap in non-variational regularization methods like deep equilibrium models and plug-and-play for inverse problems, establishing stability and convergence results including rates and estimates using an absolute Bregman distance.
Solving inverse problems \(Ax = y\) is central to a variety of practically important fields such as medical imaging, remote sensing, and non-destructive testing. The most successful and theoretically best-understood method is convex variational regularization, where approximate but stable solutions are defined as minimizers of \( \|A(\cdot) - y^δ\|^2 / 2 + α\mathcal{R}(\cdot)\), with \(\mathcal{R}\) a regularization functional. Recent methods such as deep equilibrium models and plug-and-play approaches, however, go beyond variational regularization. Motivated by these innovations, we introduce implicit non-variational (INV) regularization, where approximate solutions are defined as solutions of \(A^*(A x - y^δ) + αR(x) = 0\) for some regularization operator \(R\). When the regularization operator is the gradient of a functional, INV reduces to classical variational regularization. However, in methods like DEQ and PnP, \(R\) is not a gradient field, and the existing theoretical foundation remains incomplete. To address this, we establish stability and convergence results in this broader setting, including convergence rates and stability estimates measured via a absolute Bregman distance.