Multi-objective parameter configuration of machine learning algorithms using model-based optimization.
Daniel HornBernd BischlPublished in: SSCI (2016)
Keyphrases
- machine learning algorithms
- multi objective
- optimization algorithm
- multiple objectives
- benchmark data sets
- multiobjective optimization
- learning algorithm
- multi objective optimization
- learning problems
- evolutionary optimization
- machine learning
- decision trees
- machine learning methods
- conflicting objectives
- optimum design
- predictive accuracy
- evolutionary algorithm
- random forests
- machine learning approaches
- objective function
- machine learning systems
- machine learning models
- optimization method
- learning tasks
- nsga ii
- input features
- optimization problems
- learning models
- multi objective evolutionary algorithms
- optimal configuration
- statistical machine learning
- machine learning and data mining algorithms
- multiclass classification
- real world
- genetic algorithm
- standard machine learning algorithms
- differential evolution