English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In a computer-aided engineering design optimization problem that involvesnotoriously complex and time-consuming simulator, the prevalent approach is toreplace these simulations with a data-driven surrogate that approximates thesimulator's behavior at a much cheaper cost. The main challenge in creating aninexpensive data-driven surrogate is the generation of a sheer number of datausing these computationally expensive numerical simulations. In such cases,Active Learning (AL) methods have been used that attempt to learn aninput--output behavior while labeling the fewest samples possible. The currenttrend in AL for a regression problem is dominated by the Bayesian frameworkthat needs training an ensemble of learning models that makes surrogatetraining computationally tedious if the underlying learning model is DeepNeural Networks (DNNs). However, DNNs have an excellent capability to learnhighly nonlinear and complex relationships even for a very high dimensionalproblem. To leverage the excellent learning capability of deep networks alongwith avoiding the computational complexity of the Bayesian paradigm, in thiswork we propose a simple and scalable approach for active learning that worksin a student-teacher manner to train a surrogate model. By using this proposedapproach, we are able to achieve the same level of surrogate accuracy as theother baselines like DBAL and Monte Carlo sampling with up to 40 % fewersamples. We empirically evaluated this method on multiple use cases includingthree different engineering design domains:finite element analysis,computational fluid dynamics, and propeller design.

Data efficient surrogate modeling for engineering design: Ensemble-free batch mode deep active learning for regression