PARCO: Parallel AutoRegressive Models for Multi-Agent Combinatorial Optimization
This addresses coordination and efficiency challenges in multi-agent systems for combinatorial optimization, representing an incremental improvement over existing learning-based methods.
The paper tackles multi-agent combinatorial optimization problems by proposing PARCO, a reinforcement learning framework that integrates transformer-based communication, parallel decision-making, and conflict handling, achieving strong generalization and computational efficiency in vehicle routing and scheduling tasks.
Combinatorial optimization problems involving multiple agents are notoriously challenging due to their NP-hard nature and the necessity for effective agent coordination. Despite advancements in learning-based methods, existing approaches often face critical limitations, including suboptimal agent coordination, poor generalization, and high computational latency. To address these issues, we propose PARCO (Parallel AutoRegressive Combinatorial Optimization), a general reinforcement learning framework designed to construct high-quality solutions for multi-agent combinatorial tasks efficiently. To this end, PARCO integrates three key novel components: (1) transformer-based communication layers to enable effective agent collaboration during parallel solution construction, (2) a multiple pointer mechanism for low-latency, parallel agent decision-making, and (3) priority-based conflict handlers to resolve decision conflicts via learned priorities. We evaluate PARCO in multi-agent vehicle routing and scheduling problems, where our approach outperforms state-of-the-art learning methods, demonstrating strong generalization ability and remarkable computational efficiency. We make our source code publicly available to foster future research: https://github.com/ai4co/parco.