The state-of-the-art coding schemes for topological interference management (TIM) problems are usually handcrafted for specific families of network topologies, relying critically on experts' domain knowledge. This inevitably restricts the potential wider applications to wireless communication systems, due to the limited generalizability. This work makes the first attempt to advocate a novel intelligent coding approach to mimic topological interference alignment (IA) via local graph coloring algorithms, leveraging the new advances of graph neural networks (GNNs) and reinforcement learning (RL). The proposed LCG framework is then generalized to discover new IA coding schemes, including one-to-one vector IA and subspace IA. The extensive experiments demonstrate the excellent generalizability and transferability of the proposed approach, where the parameterized GNNs trained by small size TIM instances are able to work well on new unseen network topologies with larger size.

翻译：针对拓扑干扰管理问题的先进编码方案通常针对特定网络拓扑族手工设计，严重依赖领域专家的专业知识。由于泛化能力有限，这不可避免地限制了其在无线通信系统中的潜在广泛应用。本文首次尝试提出一种新颖的智能编码方法，利用图神经网络和强化学习的最新进展，通过局部图着色算法模拟拓扑干扰对齐。所提出的LCG框架被进一步推广以发现新的干扰对齐编码方案，包括一对一矢量干扰对齐和子空间干扰对齐。大量实验证明了所提方法的优异泛化性和可迁移性，即通过小规模拓扑干扰管理实例训练的参量化图神经网络能够在新的大规模未见网络拓扑上表现良好。