Algorithm Performance Dataset from NASA Open-Source Software
United States: NASA Center for Aerospace Information (CASI), 2020
Online
report
Zugriff:
NASA Langley Research Center has recently developed and released the open-source software Multi Model Monte Carlo with Python (MXMCPy- LAR-19756-1) as a general capability for computing the statistics of outputs from an expensive, high-fidelity model by leveraging faster, low-fidelity models for speedup. Given a fixed computational budget and a collection of models with varying cost/accuracy, multi model Monte Carlo (MC) seeks a sample allocation strategy across the models that results in an estimator with optimal variance reduction. MXMCPy is a versatile tool that enables convenient access to many existing multi-model MC approaches (over a dozen algorithms available) within one modular and extensible package [1]. With MXMCPy, users can easily compare existing methods to determine the best choice for their particular problem,while developers have a basis for implementing and sharing new variance reduction approaches. However,there is currently very little understanding about which algorithm will perform best for a given problem (defined by the correlation between and relative cost of the available models) without a brute force search.
Titel: |
Algorithm Performance Dataset from NASA Open-Source Software
|
---|---|
Autor/in / Beteiligte Person: | Geoffrey F Bomarito ; Leser, Patrick E. ; Warner, James E. ; William P Leser |
Link: | |
Veröffentlichung: | United States: NASA Center for Aerospace Information (CASI), 2020 |
Medientyp: | report |
Schlagwort: |
|
Sonstiges: |
|