The Road to VEGAS: Guiding the Search over Neutral Networks
This addresses optimization challenges in neutral networks for researchers, but it is incremental as it builds on existing methods for specific problem types.
The paper tackled optimization problems with neutrality by proposing VEGAS, a method that considers entire neutral networks and uses evolvability and multi-armed bandits to escape plateaus, showing improved results on NK-landscapes with neutrality.
VEGAS (Varying Evolvability-Guided Adaptive Search) is a new methodology proposed to deal with the neutrality property of some optimization problems. ts main feature is to consider the whole neutral network rather than an arbitrary solution. Moreover, VEGAS is designed to escape from plateaus based on the evolvability of solution and a multi-armed bandit. Experiments are conducted on NK-landscapes with neutrality. Results show the importance of considering the whole neutral network and of guiding the search cleverly. The impact of the level of neutrality and of the exploration-exploitation trade-off are deeply analyzed.