LLM-OptiRA: LLM-Driven Optimization of Resource Allocation for Non-Convex Problems in Wireless Communications

2025-05-04

AI Summary

Non-convex resource allocation problems in wireless communications are notoriously difficult to solve analytically or numerically due to their inherent structural complexity. Method: This paper introduces the first large language model (LLM)-driven framework for automatic problem reformulation and solution. Leveraging GPT-4, the method identifies non-convex structures via symbolic reasoning and constraint rewriting, transforming the original problem into an equivalent, tractable form. It further incorporates error correction and feasibility verification to minimize reliance on domain expertise. Contribution/Results: Experimental evaluation demonstrates a 96% task execution rate and an 80% success rate on GPT-4, significantly outperforming conventional optimization baselines. To the best of our knowledge, this is the first work achieving end-to-end automated modeling and solving of non-convex communication optimization problems using LLMs—establishing a novel paradigm for intelligent wireless resource management.

Technology Category

• Search and Optimization: Non-convex Optimization
• Planning, Routing, and Scheduling: Planning with Language Models
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Solving non-convex resource allocation problems poses significant challenges in wireless communication systems, often beyond the capability of traditional optimization techniques. To address this issue, we propose LLM-OptiRA, the first framework that leverages large language models (LLMs) to automatically detect and transform non-convex components into solvable forms, enabling fully automated resolution of non-convex resource allocation problems in wireless communication systems. LLM-OptiRA not only simplifies problem-solving by reducing reliance on expert knowledge, but also integrates error correction and feasibility validation mechanisms to ensure robustness. Experimental results show that LLM-OptiRA achieves an execution rate of 96% and a success rate of 80% on GPT-4, significantly outperforming baseline approaches in complex optimization tasks across diverse scenarios.

Problem

• Solving non-convex resource allocation in wireless systems
• Automating transformation of non-convex problems via LLMs
• Enhancing robustness with error correction and validation

Innovation

• LLMs detect and transform non-convex components
• Automated resolution with error correction mechanisms
• High execution and success rates outperform baselines