Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
RAAL: Resource Aware Active Learning for Multifidelity Efficient Optimization
Ist Teil von
Aerospace America, 2023-06, Vol.61 (6), p.2744
Ort / Verlag
New York: American Institute of Aeronautics and Astronautics
Erscheinungsjahr
2023
Link zum Volltext
Quelle
Nexis
Beschreibungen/Notizen
Grassi et al introduce the original resource-aware active learning (RAAL) algorithm, a multifidelity Bayesian scheme to accelerate the optimization of black-box functions. At each optimization step, the RAAL procedure computes the set of best sample locations and the associated fidelity sources that maximize the information gain to acquire during the parallel/distributed evaluation of the objective function while accounting for the limited computational budget. The scheme is demonstrated for a variety of benchmark problems, and results are discussed for both single-fidelity (SF) and multifidelily (MF) settings. In particular, they observe that the RAAL strategy optimally seeds multiple points at each iteration, which allows for a major speed up of the optimization task.