Task Force on Hyper-Heuristics (2013-2018)

Welcome to the Task Force on Hyper-heuristics within the Technical Committee of Intelligent Systems and Applications at IEEE Computational Intelligence SocietyHyper-heuristics represent one of the recent emerging meta-heuristics which attracted an increasing amount of research attention. Instead of designing heuristic methods with pre-defined parameters or mechanisms, hyper-heuristics search for or learn the selections or configurations of problem specific 'low level' heuristics which are then employed on the fly to solve the problem in hand. They therefore concern the search space of heuristics rather than solutions themselves. The algorithms are thus self-adaptive, and are able to deal with a much wider range of problems without extensive development effort. In addition to employing search algorithms, the current literature has started to investigate a wider range of computational intelligence and artificial intelligence techniques including constraint satisfaction, decision support, fuzzy rules, knowledge based systems, learning, and many others, aiming to develop advanced hybrid intelligent systems. Due to their self-adaptive nature, hyper-heuristics have been successfully applied within intelligent systems to concern various real world applications including personnel scheduling, job shop scheduling, 2D/3D strip packing, routing, assembly line, timetabling, knapsack and many other complex problems.

Along with the current state-of-the-art Hyper-heuristics research development at the interface of AI, CI and OR, we aim to further promote the scope of both intelligence techniques and applications within advanced hybrid intelligent systems.