Skip to yearly menu bar Skip to main content


Poster

Convex Regularization behind Neural Reconstruction

Arda Sahiner · Morteza Mardani · Batu Ozturkler · Mert Pilanci · John M Pauly

Keywords: [ interpretability ] [ robustness ] [ inverse problems ] [ neural networks ] [ sparsity ] [ convex optimization ] [ denoising ] [ convex duality ] [ image reconstruction ] [ neural reconstruction ]


Abstract:

Neural networks have shown tremendous potential for reconstructing high-resolution images in inverse problems. The non-convex and opaque nature of neural networks, however, hinders their utility in sensitive applications such as medical imaging. To cope with this challenge, this paper advocates a convex duality framework that makes a two-layer fully-convolutional ReLU denoising network amenable to convex optimization. The convex dual network not only offers the optimum training with convex solvers, but also facilitates interpreting training and prediction. In particular, it implies training neural networks with weight decay regularization induces path sparsity while the prediction is piecewise linear filtering. A range of experiments with MNIST and fastMRI datasets confirm the efficacy of the dual network optimization problem.

Chat is not available.