Large-scale optimization : combining co-operative coevolution and fitness inheritance

Large-scale optimization, here referring mainly to problems with many design parameters remains a serious challenge for optimization algorithms. When the problem at hand does not succumb to analytical treatment (an overwhelmingly common place situation), the engineering and adaptation of stochastic black box optimization methods tends to be a favoured approach, particularly the use of Evolutionary Algorithms (EAs). In this context, many approaches are currently under investigation for accelerating performance on large-scale problems, and we focus on two of those in this research. The first is co-operative co-evolution (CC), where the strategy is to successively optimize only subsets of the design parameters at a time, keeping the remainder fixed, with an organized approach to managing and reconciling these subspace optimization. The second is fitness inheritance (FI), which is essentially a very simple surrogate model strategy, in which, with some probability, the fitness of a solution is simply guessed to be a simple function of the finesses of that solution’s parents. Both CC and FI have been found successful on nontrivial and multiple test cases, and they use fundamentally distinct strategies. In this thesis, we explored the extent to which both of these strategies can be used to provide additional benefits. In addition to combining CC and FI, this thesis also introduces a new FI scheme which further improves the performance of CC-FI. We show that the new algorithm CC-FI is highly effective for solving problems, especially when the new FI scheme is used. In the thesis, we also explored two basic adaptive parameter setting strategies for the FI component. We found that engineering FI (and CC, where it was otherwise not present) into these algorithms led to good performance and results