论文标题
驯服深层概率预测的长尾巴
Taming the Long Tail of Deep Probabilistic Forecasting
论文作者
论文摘要
在许多应用中，从天气预后，到电力消耗估计到自动驾驶汽车轨迹预测，深层概率的预测正在引起人们的注意。但是，现有的方法集中于对最常见情况的改进，而无需解决罕见和困难案例的绩效。在这项工作中，我们在最先进的深度学习方法的概率预测方面确定了长期的尾巴行为。我们提出了两个基于力矩的尾巴测量概念，以提高困难的尾巴示例的性能：帕累托丢失和峰度丢失。峰度损失是对称测量的，作为关于损失分布平均值的第四刻。帕累托损失是不对称测量的右尾巴，使用广义帕累托分布（GPD）对损失进行建模。我们在几个现实世界中的数据集（包括时间序列和时空轨迹）上展示了方法的性能，从而在尾巴示例上取得了重大改进。
Deep probabilistic forecasting is gaining attention in numerous applications ranging from weather prognosis, through electricity consumption estimation, to autonomous vehicle trajectory prediction. However, existing approaches focus on improvements on the most common scenarios without addressing the performance on rare and difficult cases. In this work, we identify a long tail behavior in the performance of state-of-the-art deep learning methods on probabilistic forecasting. We present two moment-based tailedness measurement concepts to improve performance on the difficult tail examples: Pareto Loss and Kurtosis Loss. Kurtosis loss is a symmetric measurement as the fourth moment about the mean of the loss distribution. Pareto loss is asymmetric measuring right tailedness, modeling the loss using a generalized Pareto distribution (GPD). We demonstrate the performance of our approach on several real-world datasets including time series and spatiotemporal trajectories, achieving significant improvements on the tail examples.
