Adversarial inverse reinforcement learning (IRL) for multi-agent task allocation (MATA) is challenged by non-stationary interactions and high-dimensional coordination. Unconstrained reward inference in these settings often leads to high variance and poor generalization. We propose an attention-structured adversarial IRL framework that constrains reward inference via spatiotemporal representation learning. Our method employs multi-head self-attention (MHSA) for long-range temporal dependencies and graph attention networks (GAT) for agent-task relational structures. We formulate reward inference as a low-capacity, adaptive linear transformation of the environment reward, ensuring stable and interpretable guidance. This framework decouples reward inference from policy learning and optimizes the reward model adversarially. Experiments on benchmark MATA scenarios show that our approach outperforms representative MARL baselines in convergence speed, cumulative rewards, and spatial efficiency. Results demonstrate that attention-guided, capacity-constrained reward inference is a scalable and effective mechanism for stabilizing adversarial IRL in complex multi-agent systems.

翻译：多智能体任务分配（MATA）中的对抗性逆向强化学习（IRL）面临着非平稳交互与高维协调的挑战。在此类场景中，无约束的奖励推断通常会导致高方差与泛化性能不佳。我们提出了一种注意力结构化的对抗性IRL框架，通过时空表征学习来约束奖励推断。该方法采用多头自注意力（MHSA）捕获长程时间依赖，并利用图注意力网络（GAT）建模智能体-任务关系结构。我们将奖励推断形式化为环境奖励的低容量自适应线性变换，从而确保稳定且可解释的引导。该框架将奖励推断与策略学习解耦，并以对抗方式优化奖励模型。在基准MATA场景上的实验表明，我们的方法在收敛速度、累积奖励和空间效率方面均优于代表性的多智能体强化学习基线。结果证明，注意力引导且容量受限的奖励推断是一种可扩展且有效的机制，能够稳定复杂多智能体系统中的对抗性IRL。