Poster 130: Deep Learning-Based Feature-Aware Data Modeling for Complex Physics Simulations

Authors: Qun Liu (Louisiana State University), Subhashis Hazarika (Ohio State University), John M. Patchett (Los Alamos National Laboratory), James P. Ahrens (Los Alamos National Laboratory), Ayan Biswas (Los Alamos National Laboratory)

Abstract: Data modeling and reduction for in situ is important. Feature-driven methods for in situ data analysis and reduction are a priority for future exascale machines as there are currently very few such methods. We investigate a deep-learning-based workflow that targets in situ data processing using autoencoders. We employ integrated skip connections to obtain higher performance compared to the existing autoencoders. Our experiments demonstrate the initial success of the proposed framework and create optimism for the in situ use case.

Best Poster Finalist (BP): no

Poster: PDF
Poster summary: PDF


Back to Poster Archive Listing