Tag: surrogate modeling

A physics-aware, uncertainty-guided deep learning surrogate has been shown to replace expensive circuit simulations when optimizing ultra-low-power amplifiers for biosignal ...

A new two-stage reduced order model combines hyper-reduction and Gaussian process regression to cut both training and prediction costs for ...

Machine learning models trained on finite element simulations predicted the photoionization cross section of a donor impurity in a GaAs/AlGaAs ...

A head-to-head comparison of VAEs, DCGANs and DDPMs on simulated and experimental cylinder wake flows finds the DCGAN offers the ...

Researchers trained a machine learning surrogate on 120 simulations that predicts WAAM thermal histories in milliseconds and calibrates uncertain model ...

A PLOS One study combines experiments, electromagnetic-thermal simulation, and a neural network to evaluate an induction self-heating insert with a ...

Researchers trained Levenberg–Marquardt neural networks on bvp4c solutions to rapidly predict coupled MHD nanofluid transport between porous disks.

A new study shows that a conventional Random Forest, adjusted with an upper-quantile correction, outperformed boosting and deep tabular models ...

Researchers trained a Kolmogorov-Arnold neural network on numerical simulations of Boger nanofluid flow over a Riga surface, achieving coefficients of ...

A critical review finds that dynamic stall, vortex shedding and blade-wake interaction simultaneously govern the power, cyclic loading and noise ...