A multi-fidelity active learning method for global design optimization problems with noisy evaluations.
Riccardo PellegriniJeroen WackersRiccardo BrogliaAndrea SeraniMichel VisonneauMatteo DiezPublished in: Eng. Comput. (2023)
Keyphrases
- experimental evaluation
- active learning
- cost function
- optimization problems
- objective function
- high accuracy
- clustering method
- detection method
- dynamic programming
- prior knowledge
- genetic algorithm
- significant improvement
- user interface
- classification method
- support vector machine
- computational complexity
- synthetic data
- optimization method
- learning strategies
- combinatorial optimization
- metaheuristic
- noisy data
- optimization algorithm
- semi supervised learning
- text classification
- input data
- simulated annealing
- semi supervised
- knn
- probabilistic model
- preprocessing
- case study