Solving inverse problems, such as parameter estimation and optimal control, is a vital part of science. Many experiments repeatedly collect data and rely on machine learning algorithms to quickly infer solutions to the associated inverse problems. We find that state-of-the-art training techniques are not well-suited to many problems that involve physical processes. The highly nonlinear behavior, common in physical processes, results in strongly varying gradients that lead first-order optimizers like SGD or Adam to compute suboptimal optimization directions. We propose a novel hybrid training approach that combines higher-order optimization methods with machine learning techniques. We take updates from a scale-invariant inverse problem solver and embed them into the gradient-descent-based learning pipeline, replacing the regular gradient of the physical process. We demonstrate the capabilities of our method on a variety of canonical physical systems, showing that it yields significant improvements on a wide range of optimization and learning problems.
3-D Magnetotelluric Deep Learning Inversion Guided by Pseudo-Physical Information
Key Laboratory of Earth Exploration and Information Techniques of Ministry of Education, College of Geophysics, Chengdu University of Technology(教育部地球探测与信息技术重点实验室,成都理工大学地球物理学院);College of Computer Science and Cyber Security, Chengdu University of Technology(成都理工大学计算机科学与网络security学院)
Physical-State-Guided Diffusion Sampling for Full-Waveform Inversion
School of Mathematical Sciences, Shanghai Jiao Tong University(上海交通大学数学科学学院);CMA-Shanghai, Shanghai Jiao Tong University(上海交通大学CMA-上海);School of Mathematics and Statistics, Lanzhou University(兰州大学数学与统计学院)
PIS: A Generalized Physical Inversion Solver for Arbitrary Sparse Observations via Set Conditioned Flow Matching
School of Information, University of California, Berkeley, USA(信息学院,加州大学伯克利分校);College of Civil Engineering, Tongji University, Shanghai, China(土木学院,同济大学);Institute of Mechanical Sciences(机械科学研究院)
Full waveform inversion method based on diffusion model
School of Information and Mathematics, Yangtze University(信息与数学学院,长江大学);School of Electronic Information and Electrical Engineering, Yangtze University(电子信息与电气工程学院,长江大学)
Optical Inversion and Spectral Unmixing of Spectroscopic Photoacoustic Images with Physics-Informed Neural Networks
Georgia Institute of Technology(佐治亚理工学院);Helmholtz Zentrum Munich(慕尼黑海德堡研究中心);Technische Universität München(慕尼黑技术大学);Emory University School of Medicine(埃默里大学医学院);The Wallace H. Coulter Department of Biomedical Engineering(沃克生物医学工程系)
Including Physics in Deep Learning -- An example from 4D seismic pressure saturation inversion
Danish Hydrocarbon Research and Technology Centre(丹麦油气研究与技术中心);Technical University of Denmark(丹麦技术大学);Institute of Petroleum Engineering(石油工程学院);Heriot-Watt University(赫瑞-瓦特大学);DHRTC(丹麦油气研究与技术中心)
Physics-informed waveform inversion using pretrained wavefield neural operators
Physical Science and Engineering division, King Abdullah University of Science and Technology(国王 Abdullah 科学与技术大学物理科学与工程部门);School of Engineering and Applied Science, University of Pennsylvania(宾夕法尼亚大学工程与应用科学学院)
Physics informed Transformer-VAE for biophysical parameter estimation: PROSAIL model inversion in Sentinel-2 imagery
African Institute for Mathematical Sciences (AIMS), Accra, Ghana(非洲数学科学研究所(AIMS));African Institute for Mathematical Sciences (AIMS), Cape Town, South Africa(非洲数学科学研究所(AIMS))
Accelerating physics-informed neural networks for full waveform inversion using a hybrid quantum-classical finite-basis architecture
Department of Geophysics, Colorado School of Mines(地质学系,科罗拉多矿业学院);Department of Energy Science and Engineering, Stanford University(能源科学与工程系,斯坦福大学);Department of Petroleum Engineering, Colorado School of Mines(石油工程系,科罗拉多矿业学院)