papers
Publications (6)
cs.LG2019
A Projectional Ansatz to Reconstruction
Sören Dittmer, Peter Maass
The paper proposes a projectional framework for solving inverse problems that integrates learned and hand‑crafted priors while preserving data consistency, implemented via plug‑and…
#inverse problems#projectional methods#plug-and-play priors#regularization by denoising
stat.ML2017
Deep Learning for Tumor Classification in Imaging Mass Spectrometry
Jens Behrmann, Christian Etmann, Tobias Boskamp +3
eess.IV2018
Joint bi-modal image reconstruction of DOT and XCT with an extended Mumford-Shah functional
Di He, Ming Jiang, Alfred K. Louis +2
cs.LG2019
Singular Values for ReLU Layers
Sören Dittmer, Emily J. King, Peter Maass
The paper introduces ReLU singular values and Gaussian mean width as tools to analyze how ReLU activations interact with linear layers, providing theoretical insights, experimental…
#relu layers#singular values#gaussian mean width#neural network analysis
stat.ML2019
On the Connection Between Adversarial Robustness and Saliency Map Interpretability
Christian Etmann, Sebastian Lunz, Peter Maass +1
cs.LG2018
A Survey on Surrogate Approaches to Non-negative Matrix Factorization
Pascal Fernsel, Peter Maass