SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees
Abstract
Deep unfolding networks have recently gained popularity in the context of solving imaging inverse problems. However, the computational and memory complexity of data-consistency layers within traditional deep unfolding networks scales with the number of measurements, limiting their applicability to large-scale imaging inverse problems. We propose SGD-Net as a new methodology for improving the efficiency of deep unfolding through stochastic approximations of the data-consistency layers. Our theoretical analysis shows that SGD-Net can be trained to approximate batch deep unfolding networks to an arbitrary precision. Our numerical results on intensity diffraction tomography and sparse-view computed tomography show that SGD-Net can match the performance of the batch network at a fraction of training and testing complexity.
Cite
@article{arxiv.2101.09379,
title = {SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees},
author = {Jiaming Liu and Yu Sun and Weijie Gan and Xiaojian Xu and Brendt Wohlberg and Ulugbek S. Kamilov},
journal= {arXiv preprint arXiv:2101.09379},
year = {2021}
}