FastSTI: A Fast Conditional Pseudo Numerical Diffusion Model for Spatio-temporal Traffic Data Imputation

TL;DR

FastSTI introduces a rapid diffusion-based method for imputing missing spatiotemporal traffic data, significantly reducing computation time while maintaining or improving imputation quality, especially under high missing data rates.

Contribution

The paper proposes a high-order pseudo-numerical solver and variance schedule alignment to accelerate diffusion-based traffic data imputation without sacrificing accuracy.

Findings

Achieves 8.3x faster imputation than state-of-the-art methods.

Effectively imputes high-quality traffic data with 60-90% missing rates.

Operates efficiently with only six sampling steps.

Abstract

High-quality spatiotemporal traffic data is crucial for intelligent transportation systems (ITS) and their data-driven applications. Inevitably, the issue of missing data caused by various disturbances threatens the reliability of data acquisition. Recent studies of diffusion probability models have demonstrated the superiority of deep generative models in imputation tasks by precisely capturing the spatio-temporal correlation of traffic data. One drawback of diffusion models is their slow sampling/denoising process. In this work, we aim to accelerate the imputation process while retaining the performance. We propose a fast conditional diffusion model for spatiotemporal traffic data imputation (FastSTI). To speed up the process yet, obtain better performance, we propose the application of a high-order pseudo-numerical solver. Our method further revs the imputation by introducing a predefined alignment strategy of variance schedule during the sampling process. Evaluating FastSTI on two types of real-world traffic datasets (traffic speed and flow) with different missing data scenarios proves its ability to impute higher-quality samples in only six sampling steps, especially under high missing rates (60\% $\sim$ 90\%). The experimental results illustrate a speed-up of $\textbf{8.3} \times$ faster than the current state-of-the-art model while achieving better performance.