Gene Expression Programming: A New Adaptive Algorithm for Solving Problems

Title: Gene Expression Programming: A New Adaptive Algorithm for Solving Problems

Research Question: Can a new genetic algorithm called Gene Expression Programming (GEP) be more efficient and adaptive than existing algorithms like Genetic Algorithms (GAs) and Genetic Programming (GP)?

Methodology: The study introduces GEP, a genetic algorithm that encodes individuals as linear strings of fixed length called chromosomes. These chromosomes are then expressed as nonlinear entities of different sizes and shapes. The algorithm uses mutation, transposition, gene transposition, gene recombination, and one- and two-point recombination to modify the chromosomes. The study demonstrates the efficiency and adaptability of GEP by applying it to various problems such as symbolic regression, sequence induction with and without constant creation, block stacking, cellular automata rules for the density-classification problem, and two problems of boolean concept learning: the 11-multiplexer and the GP rule problem.

Results: The results show that GEP performs with high efficiency and surpasses the adaptivity of existing algorithms like GAs and GP. The study also demonstrates that GEP can create efficient and accurate solutions to a wide range of problems.

Implications: The study suggests that GEP is a promising new genetic algorithm that can solve problems more efficiently and adaptively than existing algorithms. Its ability to separate genome and phenotype allows for more efficient modification and evolution of individuals, making it a potentially powerful tool for solving complex problems in various fields.

Link to Article: https://arxiv.org/abs/0102027v3 Authors: arXiv ID: 0102027v3