Conditional behavior prediction (CBP) builds up the foundation for a coherent interactive prediction and planning framework that can enable more efficient and less conservative maneuvers in interactive scenarios. In CBP task, we train a prediction model approximating the posterior distribution of target agents' future trajectories conditioned on the future trajectory of an assigned ego agent. However, we argue that CBP may provide overly confident anticipation on how the autonomous agent may influence the target agents' behavior. Consequently, it is risky for the planner to query a CBP model. Instead, we should treat the planned trajectory as an intervention and let the model learn the trajectory distribution under intervention. We refer to it as the interventional behavior prediction (IBP) task. Moreover, to properly evaluate an IBP model with offline datasets, we propose a Shapley-value-based metric to verify if the prediction model satisfies the inherent temporal independence of an interventional distribution. We show that the proposed metric can effectively identify a CBP model violating the temporal independence, which plays an important role when establishing IBP benchmarks.
翻译:有条件行为预测(CBP)为协调一致的互动预测和规划框架奠定了基础,从而能够在互动情景中进行更有效和更保守的操作。在CBP任务中,我们培训了一个预测模型,以目标物剂的未来轨迹为条件,对目标物剂未来轨迹的后向分布进行适当的排序。然而,我们争辩说,CBP可能会对自主物剂如何影响目标物剂的行为产生过于自信的预期。因此,计划者质疑CBP模式的风险很大。相反,我们应该将计划轨迹视为一种干预,让模型学习干预下的轨迹分布。我们把它称为干预行为预测(IMBP)任务。此外,为了适当评价带有离线数据集的IMBP模型,我们建议采用一个基于 " 损耗值 " 的衡量标准,以核实预测模型是否满足干预物分布固有的时间独立性。我们表明,拟议的指标可以有效地确定CBP模式违反时间独立性,在建立IMB基准时起重要作用。