Efficient probabilistic surrogate modeling techniques for partially-observed large-scale dynamical systems
It addresses the challenge of efficient probabilistic surrogate modeling for dynamical systems, which is incremental as it compares existing techniques rather than introducing a new paradigm.
This paper tackles the problem of forecasting partially-observed large-scale dynamical systems, such as those described by partial differential equations, by comparing extensions to flow matching that reduce sampling steps, including direct distillation and adversarial diffusion distillation, and demonstrates results on challenging systems with direct prediction of 2D slices for 3D simulations.
This paper is concerned with probabilistic techniques for forecasting dynamical systems described by partial differential equations (such as, for example, the Navier-Stokes equations). In particular, it is investigating and comparing various extensions to the flow matching paradigm that reduce the number of sampling steps. In this regard, it compares direct distillation, progressive distillation, adversarial diffusion distillation, Wasserstein GANs and rectified flows. Moreover, experiments are conducted on a set of challenging systems. In particular, we also address the challenge of directly predicting 2D slices of large-scale 3D simulations, paving the way for efficient inflow generation for solvers.