A Bayesian approach to quantifying uncertainties and improving generalizability in traffic prediction models

TL;DR

This paper introduces a Bayesian recurrent neural network with spectral normalization for traffic prediction, providing uncertainty estimates and improved generalization across diverse datasets, crucial for traffic management.

Contribution

It proposes a novel Bayesian RNN framework with spectral normalization, enhancing uncertainty quantification and generalization in traffic prediction models.

Findings

Spectral normalization improves uncertainty estimates.

The proposed model outperforms layer normalization and non-normalized models.

Enhanced generalization on out-of-distribution traffic data.

Abstract

Deep-learning models for traffic data prediction can have superior performance in modeling complex functions using a multi-layer architecture. However, a major drawback of these approaches is that most of these approaches do not offer forecasts with uncertainty estimates, which are essential for traffic operations and control. Without uncertainty estimates, it is difficult to place any level of trust to the model predictions, and operational strategies relying on overconfident predictions can lead to worsening traffic conditions. In this study, we propose a Bayesian recurrent neural network framework for uncertainty quantification in traffic prediction with higher generalizability by introducing spectral normalization to its hidden layers. In our paper, we have shown that normalization alters the training process of deep neural networks by controlling the model's complexity and reducing the risk of overfitting to the training data. This, in turn, helps improve the generalization performance of the model on out-of-distribution datasets. Results demonstrate that spectral normalization improves uncertainty estimates and significantly outperforms both the layer normalization and model without normalization in single-step prediction horizons. This improved performance can be attributed to the ability of spectral normalization to better localize the feature space of the data under perturbations. Our findings are especially relevant to traffic management applications, where predicting traffic conditions across multiple locations is the goal, but the availability of training data from multiple locations is limited. Spectral normalization, therefore, provides a more generalizable approach that can effectively capture the underlying patterns in traffic data without requiring location-specific models.