Cost-effective search for lower-error region in material parameter space using multifidelity Gaussian process modeling.
Shion TakenoYuhki TsukadaHitoshi FukuokaToshiyuki KoyamaMotoki ShigaMasayuki KarasuyamaPublished in: CoRR (2020)
Keyphrases
- cost effective
- parameter space
- gaussian process
- hyperparameters
- likelihood function
- gaussian processes
- search space
- low cost
- gaussian process models
- search algorithm
- latent space
- regression model
- gaussian process regression
- bayesian framework
- cost effectiveness
- model selection
- high dimensional
- approximate inference
- latent variables
- parameter values
- covariance function
- semi supervised
- error rate
- support vector
- incremental learning
- denoising
- closed form
- cross validation
- similarity measure