IEEE Proceedings of the 2010 world conference on computational intelligence - WCCI 2010, Congress on evolutionary computation CEC 2010 pages:2875-2882
Congress on evolutionary computation edition:12 location:Barcelona, Spain date:18-23 July 2010
A hyper-heuristic performs search over a set of other search mechanisms.
the search, it does not require any problem-dependent data. This structure
hyper-heuristics problem-independent indirect search mechanisms. In this
we propose a learning strategy to explore elite heuristic subsets for
phases of a search. For that purpose, we apply a number of hyper-heuristics
the proposed approach to a set of home care scheduling problem instances.
results show that the learning strategy increases the performance of the
different hyper-heuristics by excluding some heuristics from the heuristic
over the tested problem instances.