U-Former ODE: Fast Probabilistic Forecasting of Irregular Time Series
About
Probabilistic forecasting of irregularly sampled time series is crucial in domains such as healthcare and finance, yet it remains a formidable challenge. Existing Neural Controlled Differential Equation (Neural CDE) approaches, while effective at modelling continuous dynamics, suffer from slow, inherently sequential computation, which restricts scalability and limits access to global context. We introduce UFO (U-Former ODE), a novel architecture that seamlessly integrates the parallelizable, multiscale feature extraction of U-Nets, the powerful global modelling of Transformers, and the continuous-time dynamics of Neural CDEs. By constructing a fully causal, parallelizable model, UFO achieves a global receptive field while retaining strong sensitivity to local temporal dynamics. Extensive experiments on five standard benchmarks -- covering both regularly and irregularly sampled time series -- demonstrate that UFO consistently outperforms ten state-of-the-art neural baselines in predictive accuracy. Moreover, UFO delivers up to 15$\times$ faster inference compared to conventional Neural CDEs, with consistently strong performance on long and highly multivariate sequences.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Probabilistic time series forecasting | ETTm2 Regular (test) | Avg NCRPS0.429 | 11 | |
| Probabilistic time series forecasting | Weather Regular (test) | Avg NCRPS0.36 | 11 | |
| Probabilistic time series forecasting | ETTm2 Irregular (test) | Average NCRPS0.43 | 11 | |
| Probabilistic time series forecasting | Weather Irregular (test) | Average NCRPS0.412 | 11 | |
| Probabilistic time series forecasting | ETTm1 Regular (test) | Avg NCRPS0.271 | 11 | |
| Probabilistic time series forecasting | ETTm1 Irregular (test) | Avg NCRPS0.294 | 11 | |
| Probabilistic time series forecasting | Traffic Regular (test) | Average NCRPS0.184 | 10 | |
| Probabilistic time series forecasting | Electricity Irregular (test) | Average NCRPS0.104 | 10 | |
| Probabilistic time series forecasting | Traffic Irregular (test) | Avg NCRPS0.188 | 10 | |
| Probabilistic time series forecasting | Electricity (test) | Average NCRPS0.102 | 10 |