Primitive Agentic First-Order Optimization

TL;DR

This paper demonstrates that simple reinforcement learning agents using minimal state information can effectively optimize quadratic problems, outperforming traditional algorithms with tuned hyperparameters, and offers a new approach to complexity management in optimization.

Contribution

It introduces a novel agent-based optimization method using primitive states and RL, showing promising results in quadratic problem classes.

Findings

RL agents outperform traditional algorithms on unseen instances

Simple partial states suffice for effective optimization

Agentic approaches can manage complexity in optimization tasks

Abstract

Efficient numerical optimization methods can improve performance and reduce the environmental impact of computing in many applications. This work presents a proof-of-concept study combining primitive state representations and agent-environment interactions as first-order optimizers in the setting of budget-limited optimization. Through reinforcement learning (RL) over a set of training instances of an optimization problem class, optimal policies for sequential update selection of algorithmic iteration steps are approximated in generally formulated low-dimensional partial state representations that consider aspects of progress and resource use. For the investigated case studies, deployment of the trained agents to unseen instances of the quadratic optimization problem classes outperformed conventional optimal algorithms with optimized hyperparameters. The results show that elementary RL methods combined with succinct partial state representations can be used as heuristics to manage complexity in RL-based optimization, paving the way for agentic optimization approaches.