Preference-Based Genetic Algorithm for Solving the Bio-Inspired NK Landscape Benchmark
Christof Ferreira Torres, Sune S. Nielsen, Grégoire Danoy, Pascal Bouvry · 2015
In molecular biology, the subject of protein structure prediction is of continued interest, not only to chart the molecular map of living cells, but also to design proteins with new functions. In this work a Preference-Based Genetic Algorithm (PBGA) is proposed aiming to optimise NK Landscape based benchmarks designed and shown to mimic properties of the Inverse Folding Problem (IFP) of proteins. The proposed algorithm incorporates a weighted sum model in order to combine fitness and diversity into a single objective function scoring a set of individuals as a whole. By adjusting the sum weights, direct control of the preferred emphasis on fitness vs. diversity in the algorithm population is achieved by means of a selection scheme iteratively removing the least contributing individuals. The proposed algorithm is compared to other algorithms where better results are achieved both in terms of fitness and diversity.