Large scale traffic forecasting with gradient boosting, Traffic4cast 2022 challenge

TL;DR

This paper presents a gradient boosting approach using PCA for traffic forecasting in the Traffic4cast 2022 challenge, achieving second place by modeling graph edge behavior with scalable, fast algorithms.

Contribution

The authors introduce a simple, scalable gradient-boosted decision tree method with PCA for traffic prediction, focusing on graph edge behavior, which is a novel approach for this challenge.

Findings

Achieved second place in Traffic4cast 2022 competition

Used PCA and LightGBM for efficient traffic prediction

Demonstrated effectiveness of gradient boosting for graph-level traffic modeling

Abstract

Accurate traffic forecasting is of the utmost importance for optimal travel planning and for efficient city mobility. IARAI (The Institute of Advanced Research in Artificial Intelligence) organizes Traffic4cast, a yearly traffic prediction competition based on real-life data [https://www.iarai.ac.at/traffic4cast/], aiming to leverage artificial intelligence advances for producing accurate traffic estimates. We present our solution to the IARAI Traffic4cast 2022 competition, in which the goal is to develop algorithms for predicting road graph edge congestion classes and supersegment-level travel times. In contrast to the previous years, this year's competition focuses on modelling graph edge level behaviour, rather than more coarse aggregated grid-based traffic movies. Due to this, we leverage a method familiar from tabular data modelling -- gradient-boosted decision tree ensembles. We reduce the dimensionality of the input data representing traffic counters with the help of the classic PCA method and feed it as input to a LightGBM model. This simple, fast, and scalable technique allowed us to win second place in the core competition. The source code and references to trained model files and submissions are available at https://github.com/skandium/t4c22 .