The impact of hyper-parameter tuning for landscape-aware performance regression and algorithm selection
Anja Jankovič, Gorjan Popovski, Tome Eftimov, Carola Doerr
2021Proceedings of the Genetic and Evolutionary Computation Conference21 citationsDOIOpen Access PDF
Abstract
Automated algorithm selection and configuration methods that build on exploratory landscape analysis (ELA) are becoming very popular in Evolutionary Computation. However, despite a significantly growing number of applications, the underlying machine learning models are often chosen in an ad-hoc manner.
Topics & Concepts
Machine learningComputer scienceRandom forestDecision treeSelection (genetic algorithm)Artificial intelligenceRegressionRegression analysisModel selectionQuality (philosophy)ComputationAlgorithmMathematicsStatisticsPhilosophyEpistemologyEvolutionary Algorithms and ApplicationsAdvanced Multi-Objective Optimization AlgorithmsMetaheuristic Optimization Algorithms Research