Abstract
We present an inversion algorithm with a deep-learning-based model compression scheme. Models are described with latent parameters of a trained variational autoencoder (VAE) neural network. Given observed data, latent parameters are inverted by minimizing the data misfit cost function using the Gauss-Newton method. This inversion algorithm is tested using both synthetic and experimental datasets. We achieve a 0.87% compression rate while maintaining high-quality reconstruction. The deep neural network renders nonlinear model compression, which largely reduces the number of unknowns; hence, it has higher computational efficiency. Furthermore, various prior knowledge that is difficult to describe with rigorous forms can be incorporated into inversion through training the neural network, which mitigates the ill-posedness of the inverse problem.
Funding Information
  • National Key Research and Development Program of China (2018YFC0603604)
  • National Natural Science Foundation of China (61971263)
  • Shuimu Tsinghua Scholar Program of Tsinghua University (2021SM031)
  • Beijing Innovation Center for Future Chip, Institute for Precision Medicine, Tsinghua University
  • Biren Tech, Beijing, China

This publication has 43 references indexed in Scilit: