Yuke Xie

Listed firm suspends nearly US$80 million in interest payments and distributions while seeking restructuring discussions with creditors

Listed firm suspends nearly US$80 million in interest payments and distributions while seeking restructuring discussions with creditors.

Drop in new home prices across 70 major cities matches the previous month, which was the sharpest decline since October

Consensus among potential buyers is that prices will continue to slide and the measures may fall flat, property agent says.

Net profit rose to a record 913.6 million yuan in the first half, 6.4 per cent more than the revised interim earnings last year

Overseas expansion should accelerate with new store openings and stronger engagement with non-Chinese customers, Goldman Sachs says.

Overseas expansion should accelerate with new store openings and stronger engagement with non-Chinese customers, Goldman Sachs says

Sportswear maker could use the competition in the highly challenging market to strengthen its core brand, CEO Qian Wei says

China Resources Land’s interim profit rose 16.2 per cent to 11.9 billion yuan, while Sunac’s loss narrowed by 14.4 per cent.

Nonlinear inverse problems in seismic imaging, particularly Full Waveform Inversion (FWI), are ill-posed and face challenges such as non-uniqueness, slow convergence, and limited ability to quantify uncertainty. This thesis presents a Bayesian framework that incorporates deep generative models, specifically a Generative Adversarial Network (GAN), to provide data-driven priors that help constrain solutions to geologically plausible structures. Variational inference techniques are applied to approximate the posterior distribution, enabling scalable inversion that accounts for uncertainty while remaining consistent with the governing physical equations. To reduce the high dimensionality of the problem, the inversion is reparameterized in the latent space of pretrained generative models, which may simplify the search space and improve convergence. A GAN-based latent reparameterization is also introduced, acting as a learned preconditioner that can accelerate optimization and reduce sensitivity to local minima. In addition, a diffusion-based Bayesian regularization strategy is proposed, where generative diffusion models are integrated directly into the FWI gradient without relying on reverse sampling, thereby avoiding instability in noisy intermediate states. Numerical experiments are used to demonstrate proof of concept, indicating that the approach may provide improved reconstructions, meaningful uncertainty estimates, and more efficient computations compared to conventional methods. Overall, the results suggest that combining generative models with Bayesian inference can offer a practical direction for advancing seismic imaging and related nonlinear inverse problems.