arxivcs.NE2026-07-24
On the Runtime Analysis of Reinforcement Learning Hyper-Heuristics
Pietro S. Oliveto, Zhenyu Wang, Peizhou Wu, Mengqing Xu
Selection Hyper-heuristics (HHs) automate algorithmic design by selecting from a set of low-level heuristics which one to apply at each stage of the optimisation process. Several impressive results have been recently rigorously proven regarding the performance of selection hyper-…