How to warm-start your unfolding network
This work addresses compressed sensing for image reconstruction, presenting an incremental improvement through ensemble techniques.
The authors tackled the compressed sensing problem by developing C-DEC, an ensemble framework that combines an overparameterized unfolding network with continuation techniques and a log-cosh loss function, resulting in smoother loss landscapes and improved reconstruction and generalization performance across all tested datasets.
We present a new ensemble framework for boosting the performance of overparameterized unfolding networks solving the compressed sensing problem. We combine a state-of-the-art overparameterized unfolding network with a continuation technique, to warm-start a crucial quantity of the said network's architecture; we coin the resulting continued network C-DEC. Moreover, for training and evaluating C-DEC, we incorporate the log-cosh loss function, which enjoys both linear and quadratic behavior. Finally, we numerically assess C-DEC's performance on real-world images. Results showcase that the combination of continuation with the overparameterized unfolded architecture, trained and evaluated with the chosen loss function, yields smoother loss landscapes and improved reconstruction and generalization performance of C-DEC, consistently for all datasets.