通过谐波退火剪枝，让多智能体强化学习在资源受限设备上更高效地优化频谱分配。
Multi-Agent Actor-Critic with Harmonic Annealing Pruning for Dynamic Spectrum Access Systems
- 将渐进剪枝融入独立策略全局评价者框架，降低计算开销。
- 在高稀疏度下性能优于线性与多项式剪枝，且稳定发现更优频谱策略。
- 适合边缘计算中需低延迟、低功耗的动态频谱接入场景。
多智能体深度强化学习（MADRL）已成为优化复杂场景中去中心化决策系统的有力工具，如动态频谱接入（DSA）。然而，由于深度学习模型计算成本高，将其部署于资源受限的边缘设备仍具挑战。为此，本文提出一种新颖的稀疏递归多智能体强化学习框架，将渐进神经网络剪枝整合至独立策略全局评价者范式中。同时引入谐波退火稀疏度调度器，在大稀疏度下表现可媲美甚至超越标准线性与多项式剪枝调度器。实验表明，所提出的DSA框架在多种训练条件下均能发现更优策略，显著优于传统DSA方法、MADRL基线及现有先进剪枝技术。
原文摘要 · Abstract (English)
Multi-Agent Deep Reinforcement Learning (MADRL) has emerged as a powerful tool for optimizing decentralized decision-making systems in complex settings, such as Dynamic Spectrum Access (DSA). However, deploying deep learning models on resource-constrained edge devices remains challenging due to their high computational cost. To address this challenge, in this paper, we present a novel sparse recurrent MARL framework integrating gradual neural network pruning into the independent actor global critic paradigm. Additionally, we introduce a harmonic annealing sparsity scheduler, which achieves comparable, and in certain cases superior, performance to standard linear and polynomial pruning schedulers at large sparsities. Our experimental investigation demonstrates that the proposed DSA framework can discover superior policies, under diverse training conditions, outperforming conventional DSA, MADRL baselines, and state-of-the-art pruning techniques.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。