Large Language Models as Surrogate Models in Evolutionary Algorithms: A Preliminary Study

TL;DR

This paper explores using large language models as surrogate models in evolutionary algorithms, demonstrating their potential to evaluate solutions effectively without additional training, thus simplifying the optimization process.

Contribution

It introduces a novel approach of employing LLMs directly as surrogate models in evolutionary algorithms, bypassing traditional training procedures.

Findings

LLMs can effectively evaluate solutions in evolutionary algorithms.

Performance of LLM-based surrogates is comparable to traditional models.

Visualization results show promising potential of LLMs in optimization tasks.

Abstract

Large Language Models (LLMs) have achieved significant progress across various fields and have exhibited strong potential in evolutionary computation, such as generating new solutions and automating algorithm design. Surrogate-assisted selection is a core step in evolutionary algorithms to solve expensive optimization problems by reducing the number of real evaluations. Traditionally, this has relied on conventional machine learning methods, leveraging historical evaluated evaluations to predict the performance of new solutions. In this work, we propose a novel surrogate model based purely on LLM inference capabilities, eliminating the need for training. Specifically, we formulate model-assisted selection as a classification and regression problem, utilizing LLMs to directly evaluate the quality of new solutions based on historical data. This involves predicting whether a solution is good or bad, or approximating its value. This approach is then integrated into evolutionary algorithms, termed LLM-assisted EA (LAEA). Detailed experiments compared the visualization results of 2D data from 9 mainstream LLMs, as well as their performance on optimization problems. The experimental results demonstrate that LLMs have significant potential as surrogate models in evolutionary computation, achieving performance comparable to traditional surrogate models only using inference. This work offers new insights into the application of LLMs in evolutionary computation. Code is available at: https://github.com/hhyqhh/LAEA.git