New paper published in Structural Safety
M. Moustapha, S. Marelli and B. Sudret published a new paper on active learning reliability methods.
The paper entitled Active learning for structural reliability: Survey, general framework and benchmark reviews various methods for active learning reliability and summarizes them under a consistent framework. This modular framework, which allows one to build on-the-fly efficient active learning strategies, is used to perform an extensive benchmark. The results of this benchmark are analysed and synthetized into recommendations for practitioners. Ultimately, it is shown that surrogate models are an important tool to fully harness the potential of advanced reliability estimation algorithms.
For more information, please follow external pagethis linkcall_made for the publication and this link for the associated report on our internal archive.