OptiMUS: Optimization Modeling Using mip Solvers and large language models

Ali AhmadiTeshnizi, Wenzhi Gao, Madeleine Udell

Primary Area: applications to robotics, autonomy, planning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: LLM, AI, Optimization modeling, optimization solvers, mathematical formulation, autonomous agents

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: We use LLMs to develop an agent that formulates and solves MIP optimization problems semi-automatically

Abstract: Optimization problems are pervasive across various sectors, from manufacturing and distribution to healthcare. However, most such problems are still solved heuristically by hand rather than optimally by state-of-the-art solvers, as the expertise required to formulate and solve these problems limits the widespread adoption of optimization tools and techniques. We introduce OptiMUS, a Large Language Model (LLM)-based agent designed to formulate and solve MLIP problems from their natural language descriptions. OptiMUS is capable of developing mathematical models, writing and debugging solver code, developing tests, and checking the validity of generated solutions. To benchmark our agent, we present NLP4LP, a novel dataset of linear programming (LP) and mixed integer linear programming (MILP) problems. Our experiments demonstrate that OptiMUS is able to solve 67\% more problems compared to a basic LLM prompting strategy. The code OptiMUS and the data for NLP4LP are available at \href{https://anonymous.4open.science/r/nlp4lp-8F62/README.md}{https://anonymous.4open.science/r/nlp4lp-8F62/README.md}

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 6499