English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Solving partial differential equations with deep learning makes it possibleto reduce simulation times by multiple orders of magnitude and unlockscientific methods that typically rely on large numbers of sequentialsimulations, such as optimization and uncertainty quantification. Two of thelargest challenges of adopting scientific AI for industrial problem settings isthat training datasets must be simulated in advance and that neural networksfor solving large-scale PDEs exceed the memory capabilities of current GPUs. Weintroduce a distributed programming API in the Julia language for simulatingtraining data in parallel on the cloud and without requiring users to managethe underlying HPC infrastructure. In addition, we show that model-paralleldeep learning based on domain decomposition allows us to scale neural networksfor solving PDEs to commercial-scale problem settings and achieve above 90%parallel efficiency. Combining our cloud API for training data generation andmodel-parallel deep learning, we train large-scale neural networks for solvingthe 3D Navier-Stokes equation and simulating 3D CO2 flow in porous media. Forthe CO2 example, we simulate a training dataset based on a commercial carboncapture and storage (CCS) project and train a neural network for CO2 flowsimulation on a 3D grid with over 2 million cells that is 5 orders ofmagnitudes faster than a conventional numerical simulator and 3,200 timescheaper.

SciAI4Industry -- Solving PDEs for industry-scale problems with deep learning