Towards Feature-Free Automated Algorithm Selection for Single-Objective Continuous Black-Box Optimization

We propose a novel method for automated algorithm selection in the domain of single-objective continuous black-box optimization. In contrast to existing methods, we use convolutional neural networks as the selection apparatus which bases its decision on a so-called 'fitness map'. This fitn...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	2021 IEEE Symposium Series on Computational Intelligence (SSCI) S. 1 - 8
Hauptverfasser:	Prager, Raphael Patrick, Vinzent Seiler, Moritz, Trautmann, Heike, Kerschke, Pascal
Format:	Tagungsbericht
Sprache:	Englisch
Veröffentlicht:	IEEE 05.12.2021
Schlagworte:	Automated Algorithm Configuration Benchmark testing Black-Box Optimization CMA-ES Convolutional neural networks Deep Learning Feature-Free Gray-scale Performance gain Point cloud compression Reinforcement learning Two dimensional displays
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!