Special Issue: Reproducibility in Evolutionary Computation
Experimental research is crucial in Evolutionary Computation. The scientific method requires that empirical results are reproducible by the authors themselves and replicable by others. Computer Science in general, and Evolutionary Computation in particular, show signs of a reproducibility crisis despite their digital underpinnings. Interest in improving reproducibility in Computer Science and other empirical sciences has grown in recent years and there is a growing number of works analysing current and best practices, obstacles and guidelines, effectiveness of journal policies, etc. Reproducibility issues in the context of Evolutionary Computation have been a topic of discussion for a long time in the context of best practices for empirical research, but there are few studies analysing reproducibility in EC research and reproducibility studies themselves are extremely rare. There is room for improvement to attain the minimum standards for reproducibility encouraged in other scientific fields. Challenges for reproducibility in EC research arise from the stochastic nature of the algorithms and, sometimes, the problems, which requires multiple runs to analyse expected behavior and variance; sensitivity of the results to the computational environment, parameter settings or implementation details; and the generalizability of conclusions to different instances of the same or related problems.
The aim of this special issue is to encourage research that analyses the topic of reproducibility in EC and showcase excellent examples of both reproducible research and reproducibility studies. Within the context of EC, we include non-evolutionary metaheuristics, swarm intelligence methods, stochastic local search, and hybrids with exact methods, i.e., matheuristics. We welcome papers that deal with the topic of reproducibility with a specific focus on the EC context, either by analysing the current state of the field or providing evidence that supports best practices for authors, journals or funding bodies. Furthermore, we also welcome papers that, in addition to an original research contribution to the field of EC, go well above the current standards of reproducibility. Finally, we also welcome high-quality reproducibility studies that attempt to reproduce (by using the materials provided by the original authors) and/or replicate (by reimplementing the materials from scratch) previously published results of interest.
Please see herefor a full description and timeline.