WI '04 Proceedings of the 2004 IEEE/WIC/ACM International Conference on Web Intelligence
Bayesian network modeling for evolutionary genetic structures
Computers & Mathematics with Applications
Guarding against premature convergence while accelerating evolutionary search
Proceedings of the 12th annual conference on Genetic and evolutionary computation
Hi-index | 0.00 |
Evolutionary algorithms have been used to effectively generate solutions to artificial life problems. However, this process may take a number of generations to complete. Research to accelerate evolutionary search has been reported, yet, insights into this evolving process have not been analyzed nor why certain characteristics are more dominant than others. This paper provides a systematic and causal explanation for these findings and why certain genes are superior. We use Bayesian Networks (BNs) to learn a graphical model to represent the learning process in the Artificial Life environment. BAyesian Network ANAlysis (BANANA) is then developed, which gives visual representation of the inter-connections among these characteristics and provides information for further insight into genetic fitness.