Training Adaptive Reconstruction Networks for Blind Inverse Problems
Institut de Mathématiques de Toulouse (IMT)（图卢兹数学研究所）；Université de Toulouse（图卢兹大学）；CNRS（法国国家科学研究中心）；Institut de Recherche en Informatique de Toulouse (IRIT)（图卢兹计算机科学研究所）；Centre de Biologie Intégrative (CBI)（综合生物学中心）
Neural networks allow solving many ill-posed inverse problems with unprecedented performance. Physics informed approaches already progressively replace carefully hand-crafted reconstruction algorithms in real applications. However, these networks suffer from a major defect: when trained on a given forward operator, they do not generalize well to a different one. The aim of this paper is twofold. First, we show through various applications that training the network with a family of forward operators allows solving the adaptivity problem without compromising the reconstruction quality significantly.Second, we illustrate that this training procedure allows tackling challenging blind inverse problems.Our experiments include partial Fourier sampling problems arising in magnetic resonance imaging (MRI) with sensitivity estimation and off-resonance effects, computerized tomography (CT) with a tilted geometry and image deblurring with Fresnel diffraction kernels.
BlinDNO: A Distributional Neural Operator for Dynamical System Reconstruction from Time-Label-Free data
Department of Mathematical Sciences, Tsinghua University, Beijing, China（清华大学数学科学系）；Yau Mathematical Sciences Center, Tsinghua University, Beijing, China（清华大学姚期智信息科学研究院）；Yanqi Lake Beijing Institute of Mathematical Sciences and Applications, Beijing, China（北京燕琦湖数学科学与应用研究院）
On the Sample Complexity of Learning for Blind Inverse Problems
MaLGa - DIBRIS - Università degli studi di Genova（MaLGa - DIBRIS - 旧金山大学）；MaLGa - DIMA - Università degli studi di Genova（MaLGa - DIMA - 旧金山大学）；Italian Institute of Technology（意大利理工学院）
Position-Blind Ptychography: Viability of image reconstruction via data-driven variational inference
Department of Informatics, University of Hamburg（汉堡大学信息学院）；Center for Free-Electron Laser Science CFEL, Deutsches Elektronen-Synchrotron DESY（自由电子激光科学中心 CFEL，德意志电子同步辐射实验室）；Department of Mathematics, Bundesstr. 55, University of Hamburg（汉堡大学数学系）；CIT School, Technical University of Munich（慕尼黑技术大学 CIT 学院）；Munich Center for Machine Learning, München（慕尼黑机器学习中心）；Massachusetts Institute of Technology (MIT)（麻省理工学院）；The NSF AI Institute for Artificial Intelligence and Fundamental Interactions（国家科学基金会人工智能与基本相互作用研究院）；Helmholtz Imaging, Deutsches Elektronen-Synchrotron DESY（海德堡成像，德意志电子同步辐射实验室）
Priors learned from legacy reconstructions inherit undetectable overconfidence
Institute for Artificial Intelligence, University of Central Florida（人工智能研究所，中央佛罗里达大学）；Department of CS, University of Central Florida（计算机科学系，中央佛罗里达大学）；Department of ECE, The University of Texas at Austin（电子工程系，德克萨斯大学奥斯汀分校）
