Command Palette
Search for a command to run...
Graph Neural Rough Differential Equations for Traffic Forecasting
Jeongwhan Choi Noseong Park
Abstract
Traffic forecasting is one of the most popular spatio-temporal tasks in the field of machine learning. A prevalent approach in the field is to combine graph convolutional networks and recurrent neural networks for the spatio-temporal processing. There has been fierce competition and many novel methods have been proposed. In this paper, we present the method of spatio-temporal graph neural rough differential equation (STG-NRDE). Neural rough differential equations (NRDEs) are a breakthrough concept for processing time-series data. Their main concept is to use the log-signature transform to convert a time-series sample into a relatively shorter series of feature vectors. We extend the concept and design two NRDEs: one for the temporal processing and the other for the spatial processing. After that, we combine them into a single framework. We conduct experiments with 6 benchmark datasets and 27 baselines. STG-NRDE shows the best accuracy in all cases, outperforming all those 27 baselines by non-trivial margins.
Code Repositories
Benchmarks
| Benchmark | Methodology | Metrics |
|---|---|---|
| traffic-prediction-on-pemsd3 | STG-NRDE | 12 steps MAE: 15.50 12 steps MAPE: 14.9 12 steps RMSE: 27.06 |
| traffic-prediction-on-pemsd4 | STG-NRDE | 12 steps MAE: 19.13 12 steps MAPE: 12.68 12 steps RMSE: 30.94 |
| traffic-prediction-on-pemsd7 | STG-NRDE | 12 steps MAE: 20.45 12 steps MAPE: 8.65 12 steps RMSE: 33.73 |
| traffic-prediction-on-pemsd7-l | STG-NRDE | 12 steps MAE: 2.85 12 steps MAPE: 7.14 12 steps RMSE: 5.76 |
| traffic-prediction-on-pemsd7-m | STG-NRDE | 12 steps MAE: 2.66 12 steps MAPE: 6.68 12 steps RMSE: 5.31 |
| traffic-prediction-on-pemsd8 | STG-NRDE | 12 steps MAE: 15.32 12 steps MAPE: 8.9 12 steps RMSE: 24.72 |
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.