A lifted Bregman strategy for training unfolded proximal neural network Gaussian denoisers
This work addresses the training efficiency bottleneck for PNNs in computational imaging, offering an incremental improvement over existing methods.
The authors tackled the problem of training unfolded proximal neural networks (PNNs) more efficiently by proposing a lifted training formulation based on Bregman distances, resulting in a bespoke computational strategy that improves training efficiency for image denoising tasks, as demonstrated through numerical simulations.
Unfolded proximal neural networks (PNNs) form a family of methods that combines deep learning and proximal optimization approaches. They consist in designing a neural network for a specific task by unrolling a proximal algorithm for a fixed number of iterations, where linearities can be learned from prior training procedure. PNNs have shown to be more robust than traditional deep learning approaches while reaching at least as good performances, in particular in computational imaging. However, training PNNs still depends on the efficiency of available training algorithms. In this work, we propose a lifted training formulation based on Bregman distances for unfolded PNNs. Leveraging the deterministic mini-batch block-coordinate forward-backward method, we design a bespoke computational strategy beyond traditional back-propagation methods for solving the resulting learning problem efficiently. We assess the behaviour of the proposed training approach for PNNs through numerical simulations on image denoising, considering a denoising PNN whose structure is based on dual proximal-gradient iterations.