Abstract
Inverse problems, which involve estimating parameters from incomplete ornoisy observations, arise in various fields such as medical imaging,geophysics, and signal processing. These problems are often ill-posed,requiring regularization techniques to stabilize the solution. In this work, weemploy Flow Matching (FM), a generative framework that integrates adeterministic processes to map a simple reference distribution, such as aGaussian, to the target distribution. Our method DAWN-FM: Data-AWare andNoise-informed Flow Matching incorporates data and noise embedding, allowingthe model to access representations about the measured data explicitly and alsoaccount for noise in the observations, making it particularly robust inscenarios where data is noisy or incomplete. By learning a time-dependentvelocity field, FM not only provides accurate solutions but also enablesuncertainty quantification by generating multiple plausible outcomes. Unlikepre-trained diffusion models, which may struggle in highly ill-posed settings,our approach is trained specifically for each inverse problem and adapts tovarying noise levels. We validate the effectiveness and robustness of ourmethod through extensive numerical experiments on tasks such as imagedeblurring and tomography.