Geometric Priors for Scientific Generative Models in Inertial Confinement Fusion
This work addresses data generation challenges in inertial confinement fusion, an incremental advancement for domain-specific applications.
The authors tackled the problem of generating multimodal data for inertial confinement fusion by developing a Wasserstein autoencoder with a hyperspherical prior, which uses a projection layer to avoid inefficient sampling and incorporates scientific constraints to validate samples, achieving unspecified performance improvements.
In this paper, we develop a Wasserstein autoencoder (WAE) with a hyperspherical prior for multimodal data in the application of inertial confinement fusion. Unlike a typical hyperspherical generative model that requires computationally inefficient sampling from distributions like the von Mis Fisher, we sample from a normal distribution followed by a projection layer before the generator. Finally, to determine the validity of the generated samples, we exploit a known relationship between the modalities in the dataset as a scientific constraint, and study different properties of the proposed model.