AI 中文总结
本研究通过对北极航运航次的受控评估，发现非线性共享奖励模型优于线性基线，添加船舶特定潜在上下文反而降低性能，表明可观测特征已能解释行为差异，为安全关键领域的AI部署提供支持。
AI 中文摘要
人工智能（AI）辅助导航可帮助北极航运适应快速变化的海冰条件，但可靠部署需要可解释且对环境变化具有鲁棒性的奖励模型。逆强化学习（IRL）提供了从船舶轨迹中恢复此类奖励的框架，而近期的元逆强化学习（meta-IRL）方法引入了潜在上下文变量以捕捉行为异质性。然而，目前尚不清楚这些潜在表示是否恢复了真正隐藏的偏好，还是仅仅重新编码了观测状态中已有的信息。我们对来自202艘船舶、覆盖9个北极航运季节的3186条自动识别系统（AIS）衍生航次进行了受控评估，比较了线性共享奖励、非线性共享奖励以及基于相同非线性架构构建的潜在上下文模型。非线性奖励比线性基线的保留似然提高了50.9%，而添加船舶特定潜在上下文则使性能降低了16.5%。行为分析、上下文探针以及预先注册的隐藏特征消融实验表明，明显的船舶级差异在很大程度上可由可观测的航线和环境条件解释，而非隐藏的船舶特定因素。此外，预测准确率、航线保真度和奖励迁移产生了不同的模型排名，这表明没有单一指标足以评估学习到的奖励。这些发现促使在添加每艘船舶的潜在上下文之前，先测试观测到的航线、环境和船舶特征是否已能解释行为差异，这有助于在安全关键领域实现更可信的AI部署。
英文摘要
Artificial Intelligence (AI)-assisted navigation can help Arctic shipping adapt to rapidly changing sea-ice conditions, but reliable deployment requires reward models that are interpretable and robust to changing environments. Inverse reinforcement learning (IRL) provides a framework for recovering such rewards from vessel trajectories, while recent meta-IRL methods introduce latent context variables to capture behavioral heterogeneity. However, it remains unclear whether these latent representations recover genuinely hidden preferences or simply re-encode information already available in the observed state. We conduct a controlled evaluation on 3,186 AIS-derived voyages from 202 vessels across nine Arctic shipping seasons, comparing a linear shared reward, a nonlinear shared reward, and a latent-context model built on the same nonlinear architecture. The nonlinear reward improves held-out likelihood by 50.9% over the linear baseline, whereas adding vessel-specific latent context reduces performance by 16.5%. Behavioral analysis, context probes, and a pre-registered feature-hiding ablation show that apparent vessel-level variation is largely explained by observable route and environmental conditions rather than hidden vessel-specific factors. Moreover, predictive accuracy, route fidelity, and reward transfer yield different model rankings, demonstrating that no single metric is sufficient to evaluate learned rewards. These findings motivate testing whether the observed route, environmental, and vessel features already explain behavioral variation before adding per-vessel latent context. This supports more trustworthy AI deployment in safety-critical domains.
