aiqbench analysis
This paper introduces HeLyMARL, a novel multi-agent reinforcement learning framework designed to optimize radio resource management in dense wireless networks under strict energy and handover budgets. By embedding Lyapunov optimization into heterogeneous MARL with virtual queues, the method successfully converts constrained finite-horizon problems into unconstrained ones while ensuring uninterrupted service and balanced throughput-fairness trade-offs. The approach demonstrates superior pacing guarantees compared to traditional Lagrangian relaxation or greedy control methods in managing cumulative resource consumption within training episodes.
Heterogeneous Multi-Agent Reinforcement Learning for Radio Resource Management under Coupled Finite-Horizon Constraints
arXiv:2608.01745v1 Announce Type: new Abstract: Maximizing throughput under proportional fairness in dense wireless networks requires jointly managing user association, scheduling, base station (BS) activation, and handover control under hard finite-horizon energy and handover budgets, which induces a fundamental tension between BS-side energy management and user-side handover regulation. While multi-agent reinforcement learning (MARL) is a natural framework for such distributed sequential control, its application here faces two difficulties: finite-horizon budget constraints cannot be evaluated at each time slot, and the nonlinear proportional fairness utility admits no principled per-slot decomposition. We propose HeLyMARL, a Lyapunov-embedded heterogeneous MARL framework that resolves both via drift-plus-penalty decomposition with virtual queues. The energy and handover constraint pressures are internalized directly into a unified per-slot reward, converting the constrained finite-horizon problem into an unconstrained MARL problem. Comparison against two Lagrangian-based alternatives reveals a timescale separation: Lagrangian relaxation regulates constraints only across training episodes, whereas the virtual queues of HeLyMARL bound cumulative budget consumption at every partial horizon within an episode, a pacing guarantee beyond the reach of greedy Lyapunov-based control. Simulations show that HeLyMARL is the only method that sustains the throughput-fairness balance together with uninterrupted service throughout the horizon, outperforming conventional MARL, Lyapunov-based, and constrained MARL benchmarks without premature budget exhaustion.
Abstract:Maximizing throughput under proportional fairness in dense wireless networks requires jointly managing user association, scheduling, base station (BS) activation, and handover control under hard finite-horizon energy and handover budgets, which induces a fundamental tension between BS-side energy management and user-side handover regulation. While multi-agent reinforcement learning (MARL) is a natural framework for such distributed sequential control, its application here faces two difficulties: finite-horizon budget constraints cannot be evaluated at each time slot, and the nonlinear proportional fairness utility admits no principled per-slot decomposition. We propose HeLyMARL, a Lyapunov-embedded heterogeneous MARL framework that resolves both via drift-plus-penalty decomposition with virtual queues. The energy and handover constraint pressures are internalized directly into a unified per-slot reward, converting the constrained finite-horizon problem into an unconstrained MARL problem. Comparison against two Lagrangian-based alternatives reveals a timescale separation: Lagrangian relaxation regulates constraints only across training episodes, whereas the virtual queues of HeLyMARL bound cumulative budget consumption at every partial horizon within an episode, a pacing guarantee beyond the reach of greedy Lyapunov-based control. Simulations show that HeLyMARL is the only method that sustains the throughput-fairness balance together with uninterrupted service throughout the horizon, outperforming conventional MARL, Lyapunov-based, and constrained MARL benchmarks without premature budget exhaustion.
Machine Learning (cs.LG)
arXiv:2608.01745 [cs.LG]
(or arXiv:2608.01745v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.01745 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 294 words
Related coverage
News
Quantum Computing Report · Jul 25, 2026
Agentic-Kube: A Graph-Enhanced Multi-Agent Reinforcement Learning Framework for Multi-Objective Kubernetes Scheduling
arXiv · Aug 25, 2026
Libra: Efficient Resource Management for Agentic RL Post-Training
arXiv · Jun 3, 2026
Libra: Efficient Resource Management for Agentic RL Post-Training
arXiv · Jun 12, 2026