arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.07632cs.ITcs.LGeess.SPmath.IT
收藏

用于可扩展无线资源管理的图神经网络:结构设计与理论分析

Graph Neural Networks for Scalable Radio Resource Management: Architecture Design and Theoretical Analysis

Hong Kong University of Science and Technology(香港科技大学);Peng Cheng Lab(鹏城实验室);ShanghaiTech University(上海科技大学);The Hong Kong Polytechnic University(香港理工大学)

Yifei Shen, Yuanming Shi, Jun Zhang, Khaled B. Letaief

摘要:

深度学习最近已经成为解决无线网络中具有挑战性的无线电资源管理问题的颠覆性技术。然而,现有的工作所采用的神经网络架构遭受的可扩展性差,通用性,并缺乏可解释性。一种长期存在的提高可扩展性和泛化能力的方法是将目标任务的结构合并到神经网络架构中。在本文中,我们提出了应用图神经网络(GNNs)来解决大规模的无线电资源管理问题,有效的神经网络架构设计和理论分析的支持。具体来说,我们首先证明,无线电资源管理问题可以制定为图优化问题,享有普遍的置换等变性。然后,我们确定了一类神经网络,命名为\emph{message passing graph neural networks}(MPGNNs)。结果表明,它们不仅满足置换等变性,而且可以推广到大规模的问题,同时享有较高的计算效率。为了解释性和理论保证,我们证明了MPGNN和一类分布式优化算法之间的等价性,然后用于分析基于MPGNN的方法的性能和推广。以功率控制和波束成形为例进行的大量仿真将证明,所提出的方法以无监督的方式使用未标记的样本进行训练,在没有特定领域知识的情况下匹配甚至优于经典的基于优化的算法。值得注意的是,所提出的方法是高度可扩展的,并且可以在单个GPU上在$6$毫秒内解决具有$1000$收发器对的干扰信道中的波束成形问题。

英文摘要:

Deep learning has recently emerged as a disruptive technology to solve challenging radio resource management problems in wireless networks. However, the neural network architectures adopted by existing works suffer from poor scalability, generalization, and lack of interpretability. A long-standing approach to improve scalability and generalization is to incorporate the structures of the target task into the neural network architecture. In this paper, we propose to apply graph neural networks (GNNs) to solve large-scale radio resource management problems, supported by effective neural network architecture design and theoretical analysis. Specifically, we first demonstrate that radio resource management problems can be formulated as graph optimization problems that enjoy a universal permutation equivariance property. We then identify a class of neural networks, named \emph{message passing graph neural networks} (MPGNNs). It is demonstrated that they not only satisfy the permutation equivariance property, but also can generalize to large-scale problems while enjoying a high computational efficiency. For interpretablity and theoretical guarantees, we prove the equivalence between MPGNNs and a class of distributed optimization algorithms, which is then used to analyze the performance and generalization of MPGNN-based methods. Extensive simulations, with power control and beamforming as two examples, will demonstrate that the proposed method, trained in an unsupervised manner with unlabeled samples, matches or even outperforms classic optimization-based algorithms without domain-specific knowledge. Remarkably, the proposed method is highly scalable and can solve the beamforming problem in an interference channel with $1000$ transceiver pairs within $6$ milliseconds on a single GPU.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑