Landscape-aware Automated Algorithm Design: An Efficient Framework for Real-world Optimization

TL;DR

This paper introduces a landscape-aware framework that combines genetic programming and large language models to efficiently discover high-performance algorithms for real-world optimization, significantly reducing evaluation costs.

Contribution

It presents a novel framework that decouples algorithm discovery from costly evaluations by using landscape similarity-guided proxy functions, enabling efficient real-world optimization.

Findings

Successfully discovered high-performance algorithms with fewer evaluations.

Demonstrated effectiveness across multiple real-world problems.

Reduced computational resources needed for algorithm discovery.

Abstract

The advent of Large Language Models (LLMs) has opened new frontiers in automated algorithm design, giving rise to numerous powerful methods. However, these approaches retain critical limitations: they require extensive evaluation of the target problem to guide the search process, making them impractical for real-world optimization tasks, where each evaluation consumes substantial computational resources. This research proposes an innovative and efficient framework that decouples algorithm discovery from high-cost evaluation. Our core innovation lies in combining a Genetic Programming (GP) function generator with an LLM-driven evolutionary algorithm designer. The evolutionary direction of the GP-based function generator is guided by the similarity between the landscape characteristics of generated proxy functions and those of real-world problems, ensuring that algorithms discovered via proxy functions exhibit comparable performance on real-world problems. Our method enables deep exploration of the algorithmic space before final validation while avoiding costly real-world evaluations. We validated the framework's efficacy across multiple real-world problems, demonstrating its ability to discover high-performance algorithms while substantially reducing expensive evaluations. This approach shows a path to apply LLM-based automated algorithm design to computationally intensive real-world optimization challenges.