Theory-inspired Parameter Control Benchmarks for Dynamic Algorithm Configuration - Archive ouverte HAL Access content directly
Conference Papers Year : 2022

Theory-inspired Parameter Control Benchmarks for Dynamic Algorithm Configuration


It has long been observed that the performance of evolutionary algorithms and other randomized search heuristics can benefit from a non-static choice of the parameters that steer their optimization behavior. Mechanisms that identify suitable configurations on the fly (“parameter control”) or via a dedicated training process (“dynamic algorithm configuration”) are thus an important component of modern evolutionary computation frameworks. Several approaches to address the dynamic parameter setting problem exist, but we barely understand which ones to prefer for which applications. As in classical benchmarking, problem collections with a known ground truth can offer very meaningful insights in this context. Unfortunately, settings with well-understood control policies are very rare. One of the few exceptions for which we know which parameter settings minimize the expected runtime is the LeadingOnes problem. We extend this benchmark by analyzing optimal control policies that can select the parameters only from a given portfolio of possible values. This also allows us to compute optimal parameter portfolios of a given size. We demonstrate the usefulness of our benchmarks by analyzing the behavior of the DDQN reinforcement learning approach for dynamic algorithm configuration.
Fichier principal
Vignette du fichier
GECCO-main.pdf (1.1 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03646966 , version 1 (20-04-2022)



André Biedenkapp, Nguyen Dang, Martin Krejca, Frank Hutter, Carola Doerr. Theory-inspired Parameter Control Benchmarks for Dynamic Algorithm Configuration. GECCO '22: Genetic and Evolutionary Computation Conference, Jul 2022, Boston, United States. ⟨10.1145/3512290.3528846⟩. ⟨hal-03646966⟩
35 View
27 Download



Gmail Facebook Twitter LinkedIn More