Uniformly convex neural networks and non-stationary iterated network Tikhonov (iNETT) method

TL;DR

This paper introduces a novel non-stationary iterated network Tikhonov (iNETT) method utilizing uniformly convex neural networks to solve ill-posed inverse problems, with theoretical convergence guarantees and practical CT imaging applications.

Contribution

It develops a new framework combining uniformly convex neural networks with the iNETT method, ensuring convergence and enabling data-driven regularization without parameter estimation.

Findings

The iNETT method converges under the proposed neural network conditions.

Numerical experiments demonstrate improved image reconstruction in CT.

Concrete examples of convex neural network architectures are provided.

Abstract

We propose a non-stationary iterated network Tikhonov (iNETT) method for the solution of ill-posed inverse problems. The iNETT employs deep neural networks to build a data-driven regularizer, and it avoids the difficult task of estimating the optimal regularization parameter. To achieve the theoretical convergence of iNETT, we introduce uniformly convex neural networks to build the data-driven regularizer. Rigorous theories and detailed algorithms are proposed for the construction of convex and uniformly convex neural networks. In particular, given a general neural network architecture, we prescribe sufficient conditions to achieve a trained neural network which is component-wise convex or uniformly convex; moreover, we provide concrete examples of realizing convexity and uniform convexity in the modern U-net architecture. With the tools of convex and uniformly convex neural networks, the iNETT algorithm is developed and a rigorous convergence analysis is provided. Lastly, we show applications of the iNETT algorithm in 2D computerized tomography, where numerical examples illustrate the efficacy of the proposed algorithm.