What is HyVis?
HyVis is a system for visualising the selection of low level heuristics by hyper-heuristics within a specified problem domain, and how often they improve. HyVis enables data and its rate of change to be visualised in real time in an intuitive interface which features fully animated graphs.
What is HyFlex?
HyFlex is a java framework for the implementation and comparison of hyper-heuristics.
What is a HyperHeuristic?
A hyper-heuristic is a learning mechanism for generating or selecting low level heuristics to solve computationally difficult problems. Under selection hyper-heuristics, a given complete solution is improved iteratively using a set of low level heuristics (which are directly applied to the problem). Hyper-heuristics run for a user specified amount of time and select low level heuristics known to apply to the problem domain, these low level heuristics are chosen at random from the pool and allowed to run for a specified amount of time, after which they are stopped and the quality of the solution is measured. If the quality of the solution is better than the best known solution seen so far during the hyper-heuristics run then we remember this one instead, if the solution meets a quality threshold set before the search began then we can end the search prematurely. A heuristic allows for finding a good solution to problems for which the ideal solution cannot be found in a polynomial amount of time. The goal of a heuristic as such is not just to find the perfect solution, but to find the best solution possible in a given amount of time.