Learning and Intelligent Optimization - 9th International Conference, LION 2015. Revised Selected Papers vol:8994 pages:74-88
Lecture Notes in Computer Science
Learning and Intelligent OptimizatioN Conference edition:9 location:Lille, France date:12-15 january 2015
A simple model shows how a reasonable update scheme for the probability vector by which a hyper-heuristic chooses the next heuristic leads to neglecting useful mutation heuristics. Empirical evidence supports this on the MaxSat, TravelingSalesman, PermutationFlowshop and VehicleRoutingProblem problems. A new approach to hyper-heuristics is proposed that addresses this problem by modeling and learning hyper-heuristics by means of a hidden Markov Model. Experiments show that this is a feasible and promising approach.