Timezone: »

 
S3RP: Self-Supervised Super-Resolution and Prediction for Advection-Diffusion Process
Chulin Wang · Kyongmin Yeo · Andres Codas · Xiao Jin · Bruce Elmegreen ·

@
in Machine Learning and the Physical Sciences »

We present a super-resolution model for an advection-diffusion process with limited information. While most of the super-resolution models assume high-resolution (HR) ground-truth data in the training, in many cases such HR dataset is not readily accessible. Here, we show that a Recurrent Convolutional Network trained with physics-based regularizations is able to reconstruct the HR information without having the HR ground-truth data. Moreover, considering the ill-posed nature of a super-resolution problem, we employ the Recurrent Wasserstein Autoencoder to model the uncertainty.

Author Information

Chulin Wang (Northwestern University)
Kyongmin Yeo (IBM Research)
Andres Codas (IBM Research)
Xiao Jin (Rensselaer Polytechnic Institute)
Bruce Elmegreen (IBM Research)
(None)

More from the Same Authors