Learning to Act and Cooperate for Distributed Black-Box Consensus Optimization

TL;DR

This paper introduces LACMAS, a novel framework using large language models to guide adaptive agent behaviors and cooperation in distributed black-box consensus optimization, enhancing efficiency and solution quality.

Contribution

It presents a trajectory-driven self-design approach with LACMAS, integrating LLM guidance and adaptive dynamics for improved multi-agent optimization.

Findings

LACMAS outperforms strong baselines in solution quality.

It improves convergence and communication efficiency.

Experiments validate its effectiveness on real-world tasks.

Abstract

Distributed blackbox consensus optimization is a fundamental problem in multi-agent systems, where agents must improve a global objective using only local objective queries and limited neighbor communication. Existing methods largely rely on handcrafted update rules and static cooperation patterns, which often struggle to balance local adaptation, global coordination, and communication efficiency in heterogeneous nonconvex environments. In this paper, we take an initial step toward trajectory-driven self-design for distributed black-box consensus optimization. We first redesign the agent-level swarm dynamics with an adaptive internal mechanism tailored to decentralized consensus settings, improving the balance between exploration, convergence, and local escape. Built on top of this adaptive execution layer, we propose Learning to Act and Cooperate (LACMAS), a trajectorydriven framework in which large language models provide sparse highlevel guidance for shaping both agentinternal action behaviors and agentexternal cooperation patterns from historical optimization trajectories. We further introduce a phased cognitive scheduling strategy to activate different forms of adaptation in a resource-aware manner. Experiments on standard distributed black-box benchmarks and real-world distributed tasks show that LAC-MAS consistently improves solution quality, convergence efficiency, and communication efficiency over strong baselines, suggesting a practical route from handcrafted distributed coordination toward self-designing multi-agent optimization systems.