Does Dimensionality Reduction via Random Projections preserve Landscape Features?

Genetic and Evolutionary Computation Conference (GECCO), pp. 919–927

Abstract

Exploratory Landscape Analysis (ELA) provides numerical features for characterizing black-box optimization problems. In high-dimensional settings, however, ELA suffers from sparsity effects, high estimator variance, and the prohibitive cost of computing several feature classes. Dimensionality reduction has therefore been proposed as a way to make ELA applicable in such settings, but it remains unclear whether features computed in reduced spaces still reflect intrinsic properties of the original landscape.

Authors 4

1. Iván Olarte Rodríguez Aachen Leiden Institute of Advanced Computer Science - LIACS

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science - LIACS, Leiden University, Leiden, Netherlands

2. Anja Jankovič Aachen Chair for Artificial Intelligence

   RWTH Aachen University

   Affiliation as printed

   Chair for Artificial Intelligence, RWTH Aachen University, Aachen, Germany

3. Thomas Bäck Aachen Leiden Institute of Advanced Computer Science - LIACS

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science - LIACS, Leiden University, Leiden, Netherlands

4. Elena Raponi Aachen Leiden Institute of Advanced Computer Science - LIACS

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science - LIACS, Leiden University, Leiden, Netherlands