In this paper, we investigate the potential of using Large Language Models (LLMs) such as GPT-4 to generate novel hybrid swarm intelligence optimization algorithms. We use the LLM to identify and decompose six well-performing swarm algorithms for continuous optimization: Particle Swarm Optimization (PSO), Cuckoo Search (CS), Artificial Bee Colony (ABC), Grey Wolf Optimizer (GWO), Self-Organizing Migrating Algorithm (SOMA), and Whale Optimization Algorithm (WOA). We leverage GPT-4 to propose a hybrid algorithm that combines the strengths of these techniques for two distinct use-case scenarios. Our focus is on the process itself and various challenges that emerge during the use of GPT-4 to fulfill a series of set tasks. Furthermore, we discuss the potential impact of LLM-generated algorithms in the metaheuristics domain and explore future research directions.
This paper presents a novel multi-chaotic framework, which is used for the parent selection process in Success-History based Adaptive Differential Evolution (SHADE) algorithm. Created variant of the Differential Evolution (DE) algorithm was named MC-SHADE and its performance is tested on the CEC2014 benchmark set in order to participate in the CEC2016 competition on bound constrained single objective numerical optimization — single parameter-operator set based case.
This research paper presents a new approach to population size reduction in Success-History based Adaptive Differential Evolution (SHADE). The current L-SHADE algorithm uses fitness function value to select individuals which will be deleted from the current population. Algorithm variant proposed in this paper (Net L-SHADE) is using the information from evolutionary process to construct a network of individuals and the ones which would be deleted are selected based on their degree of centrality. The proposed technique is compared to state-of-art L-SHADE on CEC2015 benchmark set and the results are reported.
This study investigates the use of the GPT-4 Turbo, a large language model, to enhance the Self-Organizing Migrating Algorithm (SOMA), specifically its All to All variant (SOMA-ATA). Utilizing the model's extensive context capacity for iterative prompting without feedback, we sought to autonomously generate superior algorithmic versions. Contrary to our initial hypothesis, the improvements did not progress linearly. Nevertheless, one iteration stood out significantly, consistently outperforming the baseline across various pairwise comparisons and showing a robust performance profile. This iteration's structure deviated substantially from traditional SOMA principles, underscoring the potential of large language models to create distinctive and effective algorithmic strategies. The results affirm the methodology's ability to produce high-performing algorithms without expert intervention, setting the stage for future research to integrate feedback mechanisms and conduct detailed code analyses to understand further the modifications made by the Large Language Models.