Optimization Problem Solving Can Transition to Evolutionary Agentic Workflows

TL;DR

This paper advocates for replacing human-dependent optimization workflows with autonomous evolutionary agentic systems powered by foundation models, enabling scalable and adaptive problem-solving in industrial contexts.

Contribution

It introduces the concept of evolutionary agentic workflows that utilize foundation models and evolutionary search to automate optimization processes.

Findings

Case studies in cloud resource scheduling demonstrate effectiveness.

Autonomous workflows reduce reliance on human experts.

Potential to accelerate industrial adoption of advanced optimization methods.

Abstract

This position paper argues that optimization problem solving can transition from expert-dependent to evolutionary agentic workflows. Traditional optimization practices rely on human specialists for problem formulation, algorithm selection, and hyperparameter tuning, creating bottlenecks that impede industrial adoption of cutting-edge methods. We contend that an evolutionary agentic workflow, powered by foundation models and evolutionary search, can autonomously navigate the optimization space, comprising problem, formulation, algorithm, and hyperparameter spaces. Through case studies in cloud resource scheduling and ADMM parameter adaptation, we demonstrate how this approach can bridge the gap between academic innovation and industrial implementation. Our position challenges the status quo of human-centric optimization workflows and advocates for a more scalable, adaptive approach to solving real-world optimization problems.