发表机构
School of Automotive Studies, Tongji University（同济大学汽车学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对自动驾驶多智能体强化学习策略的黑箱评估问题，提出基于MCTS的信息论诊断框架，构建策略最优性评分以精准暴露策略缺陷，为内在多智能体策略质量提供严格基准标准。
AI 中文摘要
在自动驾驶领域评估多智能体强化学习（MARL）策略，本质上依赖于奖励曲线、成功率等外在统计指标，这些指标往往掩盖了内在策略退化和算法盲区。为打破黑箱评估，本文提出一种新型信息论诊断框架。我们利用完全收敛的蒙特卡洛树搜索（MCTS）作为渐近基准，建立理论真值基线分布；通过前向KL散度构建有界策略最优性评分（$\boldsymbol{\textit{M}}_{\textit{opt}}$），严格惩罚致命协作遗漏。关键是，我们将该指标在语义上解耦为横向和纵向维度，形成精细的“语义显微镜”。对最先进的MARL架构和探索机制开展的广泛时空诊断表明，该框架可明确揭示隐藏的方向偏差、识别时间平均策略陷阱，并将启发式超参数调优转化为可视觉追踪的轨迹优化，为基准测试内在多智能体策略质量建立了严格的模型无关标准。
英文摘要
Evaluating Multi-Agent Reinforcement Learning (MARL) policies in autonomous driving fundamentally relies on extrinsic statistical indicators (e.g., reward curves and success rates), which often mask intrinsic policy degradation and algorithmic blind spots. To break this black-box evaluation, this letter proposes a novel information-theoretic diagnostic framework. By leveraging a fully converged Monte Carlo Tree Search (MCTS) as an asymptotic oracle, we establish a theoretical ground-truth baseline distribution. We formulate a bounded policy optimality score ($\mathcal{M}_{opt}$) using the forward KL divergence to rigorously penalize fatal collaborative omissions. Crucially, we semantically decouple this metric into lateral and longitudinal dimensions, creating a granular "semantic microscope". Extensive spatial and temporal diagnostics on state-of-the-art MARL architectures and exploration mechanisms demonstrate that our framework conclusively exposes hidden directional biases, identifies temporal average-policy traps, and transforms heuristic hyperparameter tuning into a visually trackable trajectory optimization. This framework establishes a rigorous, model-agnostic standard for benchmarking intrinsic multi-agent policy quality.
Comments8 pages, 7 figures
