Impact of Spatial Transformations on Exploratory and Deep-Learning Based Landscape Features of CEC2022 Benchmark Suite

International Joint Conference on Computational Intelligence, pp. 60–71

Abstract

When benchmarking optimization heuristics, we need to take care to avoid an algorithm exploiting biases in the construction of the used problems. One way in which this might be done is by providing different versions of each problem but with transformations applied to ensure the algorithms are equipped with mechanisms for successfully tackling a range of problems. In this paper, we investigate several of these problem transformations and show how they influence the low-level landscape features of problems from the Congress on Evolutionary Computation 2022 benchmark suite. Our results highlight that even relatively small transformations can significantly alter the measured landscape features. This poses a wider question of what properties we want to preserve when creating problem transformations, and how to measure them fairly.

Authors 5

1. Haoran Yin Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, Netherlands, --- Select a Country ---

2. Diederick Vermetten Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, Netherlands, --- Select a Country ---

3. Furong Ye

   Chinese Academy of Sciences · Institute of Software

   Affiliation as printed

   ISCAS, Chinese Academy of Science, Beijing, China, --- Select a Country ---

4. Thomas Bäck Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, Netherlands, --- Select a Country ---

5. Anna V. Kononova Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, Netherlands, --- Select a Country ---