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

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.

Automated summary

We propose a non-stationary iterated network Tikhonov (iNETT) method for solving ill-posed inverse problems. The iNETT method uses deep neural networks to build a data-driven regularizer and eliminates the need to estimate the optimal regularization parameter. To ensure the theoretical convergence of iNETT, we introduce uniformly convex neural networks as the data-driven regularizer. We provide rigorous theories, detailed algorithms, and concrete examples of convexity and uniform convexity in neural networks, and develop the iNETT algorithm with a rigorous convergence analysis.