Searching for programs by genetic programming has been successfully applied to many areas. However, the vast and rugged fitness landscape of searching programs often limits the further improvement of genetic programming. It takes a significant amount of time for genetic programming to traverse the huge search space and...
Zhi-Xing Huang, Bing Xue, Yi Mei et al.· ACM Transactions on Evolutio...· 0 citations
Enhancing the classification performance is more challenging than reducing the number of features in multiobjective feature selection. Such imbalance might lead the population to converge to a region on the Pareto front with a small number of features. Meanwhile, increasing the classification performance is time consum...
Analysis shows that many generated operators use semantics to guide selection, suggesting that LLMs can produce non-trivial search heuristics from the task description alone, and the relationship between public LLM leaderboard rankings and GP performance is examined.