Uncertainty Quantification for Traffic Forecasting: A Unified Approach

Weizhu Qian1; Dalin Zhang1; Yan Zhao1; Kai Zheng2; James J.Q. Yu3

10.1109/ICDE55515.2023.00081

2023

1063-6382

979-8-3503-2228-6

2023 IEEE 39th International Conference on Data Engineering (ICDE)

2023-April

992-1004

3-7 April 2023

Anaheim, CA, USA

Uncertainty is an essential consideration for time series forecasting tasks. In this work, we specifically focus on quantifying the uncertainty of traffic forecasting. To achieve this, we develop Deep Spatio-Temporal Uncertainty Quantification (DeepSTUQ), which can estimate both aleatoric and epistemic uncertainty. We first leverage a spatio-temporal model to model the complex spatio-temporal correlations of traffic data. Subsequently, two independent sub-neural networks maximizing the heterogeneous log-likelihood are developed to estimate aleatoric uncertainty. For estimating epistemic uncertainty, we combine the merits of variational inference and deep ensembling by integrating the Monte Carlo dropout and the Adaptive Weight Averaging re-training methods, respectively. Finally, we propose a post-processing calibration approach based on Temperature Scaling, which improves the model’s generalization ability to estimate uncertainty. Extensive experiments are conducted on four public datasets, and the empirical results suggest that the proposed method outperforms state-of-the-art methods in terms of both point prediction and uncertainty quantification.

traffic forecasting uncertainty quantification variational inference deep ensembling model calibration

其他

EI

20233314551409

Monte Carlo methods ; Uncertainty analysis

Probability Theory:922.1 ; Mathematical Statistics:922.2

IEEE

1.Aalborg University, Denmark
2.University of Electronic Science and Technology of China, China
3.Southern University of Science and Technology, China

Weizhu Qian,Dalin Zhang,Yan Zhao,et al. Uncertainty Quantification for Traffic Forecasting: A Unified Approach[C],2023:992-1004.