... optimization multi-objective optimization Gaussian process acquisition function data-driven design design optimization machine learning metamodeling simulation-based design The distinct treatment of dominated designs adopted by AFs, such as the EEI function and the proposed approach, generates...

... as design variables. Gaussian process (GP) regression models are trained to predict the relationship between latent features and properties for property-driven optimization. The optimal structural designs are reconstructed by mapping the optimized latent feature values to the original image space. Compared...

... 2021 inverse design invertible neural network machine learning probabilistic modeling Gaussian process gas turbine aerodynamic design data-driven design design automation design methodology generative design Advanced Research Projects Agency 10.13039/100009224 DE-AR0001204...

...Liwei Wang; Suraj Yerramilli; Akshay Iyer; Daniel Apley; Ping Zhu; Wei Chen Scientific and engineering problems often require the use of artificial intelligence to aid understanding and the search for promising designs. While Gaussian processes (GP) stand out as easy-to-use and interpretable...

... representation Gaussian process topological domain conditional simulation uncertainty analysis To enable stochastic analysis (e.g., Monte Carlo simulation) of the network system, an effective way of generating random realization from the uncertainty representation model is needed. As discussed...

... classes to accommodate spatially varying desired properties. The key challenge is the lack of an inherent ordering or “distance” measure between different classes of microstructures in meeting a range of properties. To overcome this hurdle, we extend the newly developed latent-variable Gaussian process...

... and stochastic processes. By randomly sampling the time-independent random variables, multiple LSTM networks can be trained and leveraged with the Gaussian process (GP) regression to construct a global surrogate model for the time-dependent limit state function. In detail, a set of augmented data is first...

...Sangjune Bae; Chanyoung Park; Nam H. Kim An approach is proposed to quantify the uncertainty in probability of failure using a Gaussian process (GP) and to estimate uncertainty change before actually adding samples to GP. The approach estimates the coefficient of variation (CV) of failure...

... surface, y m ( x ) , and to determine the parameters of the bias correction function. The hyperparameters of the Gaussian process are briefly listed in Appendix A. The results are shown in case 1 through case 6 in Table 1 . Column 4 shows the standard deviations obtained from the bias...

... by posterior standard deviations) by an amount that ranges from minimal to substantial, depending on the characteristics of the specific responses that are combined. 27 08 2011 03 07 2012 21 09 2012 28 09 2012 multiple responses Gaussian process model updating calibration...