Accurate and efficient path loss prediction in mmWave communication plays an important role in large-scale deployment of the mmWave-based 5G mobile communication systems. Existing methods often present limitations in accuracy and efficiency and fail to fulfill the requirements of cell planning, especially in dense urban environments. In this paper, we propose a novel training method called multi-way local attentive learning, which allows for learning from multiple perspectives on the same set of training samples with local attention paid to each subset of the entire dataset. The sample data set can be partitioned in various ways with respect to different attributes, such that a larger amount of knowledge can be extracted from the same data set. The proposed scheme outperforms the existing schemes in terms of prediction accuracy at the average RMSE of 6.01 dBm.

In addressing multi-agent reinforcement learning (MARL) challenges, Multi-Agent Transformer (MAT) has demonstrated a number of notable successes. In various benchmarks, MAT has consistently showcased strong performance. A key observation in the latest MAT frameworks is the MARL modeling with sequence modeling (SM) to represent the inter-agent relationships by self-attention mechanisms. This paper applies graph-based modeling to represent the inter-agent relationships present in agent interactions for improved performance. To this end, we introduce so-called MAT-GAT model that leverages Graph Attention Networks (GAT) to allow for individualized consideration of interactions between agents. This enables a MAT to pay more attention on the information relative to inter-agent interactions within a cooperative MARL environment. To evaluate the performance of MAT-GAT, we conduct a series of benchmark tests across three different levels of the StarCraft Multi-Agent Challenge (SMAC) tasks and MuJoCo Half-cheetah task. The test results indicate that MAT-GAT outperforms both the original MAT and state-of-the-art baselines such as QMIX, particularly in complex environments. This demonstrates MAT-GAT’s improved performance with respect to the representation capabilities and learning efficiencies of MAT-GAT.