Drift Analysis with Fitness Levels for Elitist Evolutionary Algorithms
share
The fitness level method is a popular tool for analyzing the computation time of elitist evolutionary algorithms. Its idea is to divide the search space into multiple fitness levels and estimate lower and upper bounds on the computation time using transition probabilities between fitness levels. However, the lower bound generated from this method is often not tight. To improve the lower bound, this paper rigorously studies an open question about the fitness level method: what are the tightest lower and upper time bounds that can be constructed based on fitness levels? To answer this question, drift analysis with fitness levels is developed, and the tightest bound problem is formulated as a constrained multi-objective optimization problem subject to fitness level constraints. The tightest metric bounds from fitness levels are constructed and proven for the first time. Then the metric bounds are converted into linear bounds, where existing linear bounds are special cases. This paper establishes a general framework that can cover various linear bounds from trivial to best coefficients. It is generic and promising, as it can be used not only to draw the same bounds as existing ones, but also to draw tighter bounds, especially on fitness landscapes where shortcuts exist. This is demonstrated in the case study of the (1+1) EA maximizing the TwoPath function.
READ FULL TEXT