Efficient Learning of Accurate Surrogates for Simulations of Complex Systems
This work addresses the challenge of creating efficient and reliable surrogates for complex physical systems, such as nuclear matter, which is incremental as it builds on existing surrogate modeling approaches.
The paper tackles the problem of building accurate surrogates for complex simulations when data are noisy or sparse, by introducing an online learning method with optimizer-driven sampling, which outperforms traditional methods in accuracy around local extrema and reliably generates highly accurate surrogates for nuclear matter simulations with few evaluations.
Machine learning methods are increasingly used to build computationally inexpensive surrogates for complex physical models. The predictive capability of these surrogates suffers when data are noisy, sparse, or time-dependent. As we are interested in finding a surrogate that provides valid predictions of any potential future model evaluations, we introduce an online learning method empowered by optimizer-driven sampling. The method has two advantages over current approaches. First, it ensures that all turning points on the model response surface are included in the training data. Second, after any new model evaluations, surrogates are tested and "retrained" (updated) if the "score" drops below a validity threshold. Tests on benchmark functions reveal that optimizer-directed sampling generally outperforms traditional sampling methods in terms of accuracy around local extrema, even when the scoring metric favors overall accuracy. We apply our method to simulations of nuclear matter to demonstrate that highly accurate surrogates for the nuclear equation of state can be reliably auto-generated from expensive calculations using a few model evaluations.