Non-intrusive surrogate modelling using sparse random features with applications in crashworthiness analysis
This work addresses uncertainty quantification for crashworthiness analysis, offering a more efficient surrogate modeling method that is incremental in improving existing techniques.
The paper tackled efficient surrogate modeling for uncertainty quantification in crashworthiness analysis by proposing a novel approach using sparse random features with self-supervised dimensionality reduction, achieving superiority over state-of-the-art methods like Polynomial Chaos Expansions and Neural Networks on synthetic and real data.
Efficient surrogate modelling is a key requirement for uncertainty quantification in data-driven scenarios. In this work, a novel approach of using Sparse Random Features for surrogate modelling in combination with self-supervised dimensionality reduction is described. The method is compared to other methods on synthetic and real data obtained from crashworthiness analyses. The results show a superiority of the here described approach over state of the art surrogate modelling techniques, Polynomial Chaos Expansions and Neural Networks.