What is Wrong with End-to-End Learning for Phase Retrieval?
Not on the map. Why?
Abstract (arXiv)
For nonlinear inverse problems that are prevalent in imaging science, symmetries in the forward model are common. When data-driven deep learning approaches are used to solve such problems, these intrinsic symmetries can cause substantial learning difficulties. In this paper, we explain how such difficulties arise and, more importantly, how to overcome them by preprocessing the training set before any learning, i.e., symmetry breaking. We take far-field phase retrieval (FFPR), which is central to many areas of scientific imaging, as an example and show that symmetric breaking can substantially improve data-driven learning. We also formulate the mathematical principle of symmetry breaking.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 21 of its 37 references that are in the map. First posted on arXiv 18 Mar 2024.
Builds on
21 of 37 references.
References · 21
- 2018Scale-recurrent Network for Deep Image Deblurringcited by 333
- 2021Physics-informed neural networks with hard constraints for inverse designcited by 280
- 2020hp-VPINNs: Variational Physics-Informed Neural Networks With Domain Decompositioncited by 236
- 2017A Review of Convolutional Neural Networks for Inverse Problems in Imagingcited by 81
- 2022Learning two-phase microstructure evolution using neural operators and autoencoder architecturescited by 72
- 2020The gap between theory and practice in function approximation with deep neural networkscited by 54
- 2018prDeep: Robust Phase Retrieval with a Flexible Deep Networkcited by 47
- 2017Lensless computational imaging through deep learningcited by 44
- 2019Deep Non-Rigid Structure from Motioncited by 38
- 2023KBNet: Kernel Basis Network for Image Restorationcited by 33
- 2019FrameNet: Learning Local Canonical Frames of 3D Surfaces from a Single RGB Imagecited by 26
- 2018Low Photon Count Phase Retrieval Using Deep Learningcited by 24
9 more
- 2018Deep speckle correlation: a deep learning approach towards scalable imaging through scattering mediacited by 23
- 2018Real-time coherent diffraction inversion using deep generative networkscited by 17
- 2019Deep Iterative Reconstruction for Phase Retrievalcited by 14
- 2019Phase Retrieval Using Conditional Generative Adversarial Networkscited by 11
- 2020Deep NRSfM++: Towards Unsupervised 2D-3D Lifting in the Wildcited by 10
- 2020Inverse Problems, Deep Learning, and Symmetry Breakingcited by 8
- 2022SiSPRNet: End-to-End Learning for Single-Shot Phase Retrievalcited by 7
- 2022Egocentric Scene Understanding via Multimodal Spatial Rectifiercited by 7
- 2022ADMM based Fourier phase retrieval with untrained generative priorcited by 2
Built on by
Citing papers per year: 2025: 2.
2 papers.