Tag: inverse problems

Researchers have developed an importance-sampling-based black box variational inference algorithm that reuses earlier simulation evaluations to cut the computational cost ...

A physics-informed neural network framework reconstructs landslide-induced loads on buried pipelines from sparse monitoring data with errors below 0.2% in ...

A new method called physics-embedded inverse learning trains neural networks to estimate unobservable physical parameters by supervising them with final ...

Researchers have combined a phase-dependent signal preprocessing step with a dual U-Net deep learning framework to reconstruct planar antenna array ...

A Transformer-based network that treats each ultrasound sensor's time series as a token reconstructs photoacoustic images in a single forward ...

A new conditioning-controlled retrieval framework jointly derives broadband land surface temperature and emissivity from paired ground-based longwave irradiance measurements, achieving ...

Researchers have developed a fast, low-cost method that converts X-ray diffraction fingerprints of river sediment into spatially resolved erosion rate ...

A new mathematical framework establishes when complex oscillators can be uniquely tuned to multiple simultaneous targets, validated on electronic genetic ...

Researchers in Lübeck have trained a neural network to reconstruct CT images using only raw X-ray measurements and scanner physics, ...

Vietnamese researchers have trained a 3D U-Net to reconstruct facial soft tissue directly from skull geometry, achieving millimeter-level accuracy while ...