AVO inversion via invertible neural networks

What is it about?

Sun et al. propose a method to mitigate the dependence of AVO inversions on accurate training samples and initial models by solving inverse problems with the recently developed invertible neural networks (INNs).

Featured Image

Why is it important?

This method can reliably invert low to medium frequency velocities and densities with randomly generated easy-to-access datasets rather than trustworthy training samples or well-prepared initial models.