Editing Theodore C. Belding

Title: Theodore C. Belding

Main Research Question: How does an evolutionary algorithm (EA) search a fitness landscape, and on what fitness landscapes does a particular EA perform well?

Methodology: Belding extended the work done in the royal road project, a research program aimed at understanding the search capabilities of EAs. He introduced a dialectical heuristic, a method for searching for a simple class of functions that a specific EA, in this case, the simple genetic algorithm (SGA), optimizes better than other methods like hillclimbers. As an example, he compared the SGA with a set of hillclimbers on a subset of hyperplane-deﬁned functions, known as pothole functions.

Results: Belding found that the SGA performed better than the hillclimbers on these pothole functions. This suggests that the building-block hypothesis, a major tenet of traditional genetic algorithm theory, may not be as universally applicable as previously thought.

Implications: This research highlights the importance of understanding the search capabilities of EAs and the factors that influence their performance. It also provides insights into the limitations of the building-block hypothesis and suggests that more research is needed to develop better search methods for specific types of problems.

Link to Article: https://arxiv.org/abs/0104011v1
Authors: 
arXiv ID: 0104011v1

[[Category:Computer Science]]
[[Category:Research]]
[[Category:Search]]
[[Category:Functions]]
[[Category:As]]
[[Category:Belding]]