Improving Multi-agent Coordination by Learning to Estimate Contention
This addresses efficient and fair resource allocation for decentralized multi-agent systems, with incremental improvements in learning speed and deployment feasibility.
The paper tackles the problem of multi-agent coordination in large-scale systems by introducing ALMA-Learning, which uses the ALMA heuristic to achieve near-optimal (<5% loss) and fair allocations in synthetic and real-world scenarios like meeting scheduling.
We present a multi-agent learning algorithm, ALMA-Learning, for efficient and fair allocations in large-scale systems. We circumvent the traditional pitfalls of multi-agent learning (e.g., the moving target problem, the curse of dimensionality, or the need for mutually consistent actions) by relying on the ALMA heuristic as a coordination mechanism for each stage game. ALMA-Learning is decentralized, observes only own action/reward pairs, requires no inter-agent communication, and achieves near-optimal (<5% loss) and fair coordination in a variety of synthetic scenarios and a real-world meeting scheduling problem. The lightweight nature and fast learning constitute ALMA-Learning ideal for on-device deployment.