2020-01-17: Up to two billion times acceleration of scientific simulations with deep neural architecture search https://arxiv.org/abs/2001.08055v1The combined update steps from equations (??) and (??), and the use of a ranking function in assigning rewards, make DENSE a robust algorithm to simultaneously learn the weights and find the right architecture for a given problem

‹›
Computer simulations are invaluable tools for scientific discovery. However,
accurate simulations are often slow to execute, which limits their
applicability to extensive parameter exploration, large-scale data analysis,
and uncertainty quantification. A promising route to accelerate simulations by
building fast emulators with machine learning requires large training datasets,
which can be prohibitively expensive to obtain with slow simulations. Here we
present a method based on neural architecture search to build accurate
emulators even with a limited number of training data. The method successfully
accelerates simulations by up to 2 billion times in 10 scientific cases
including astrophysics, climate science, biogeochemistry, high energy density
physics, fusion energy, and seismology, using the same super-architecture,
algorithm, and hyperparameters. Our approach also inherently provides emulator
uncertainty estimation, adding further confidence in their use. We anticipate
this work will accelerate research involving expensive simulations, allow more
extensive parameters exploration, and enable new, previously unfeasible
computational discovery.