A sugeno-based search width decay schedule in the ACOR algorithm
Ashraf M. Abdelbar, Khalid M. Salama · 2017
ACORis a fairly recent Ant Colony Optimization (ACO) algorithm for continuous problem domains. ACOR's search width parameter ξ controls the extent to which the search is concentrated around the best solutions encountered so far. In previous work, we proposed a variation of ACOR, in which ACOR's search width parameter decays over time, being reduced by a fixed fraction in each iteration. In the present work, we propose a parameterized decay schedule based on Sugeno's fuzzy complement operator. We evaluate our proposal in the context of neural network training using 36 popular datasets. We find that our approach produces solutions that are better, to a statistically significant extent, than both the standard ACORalgorithm and our previously-proposed variation.