Research map / Time series / Causal and generative time series
Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use TimeGrad instead of Transformer-MAF | Multivariate probabilistic forecasting · Solar | CRPS_sum: 0.301 -> 0.287 | Helped |
| Use TimeGrad instead of Transformer-MAF | Multivariate probabilistic forecasting · Electricity | CRPS_sum: 0.0207 -> 0.0206 | Helped |
| Use TimeGrad instead of Transformer-MAF | Multivariate probabilistic forecasting · Traffic | CRPS_sum: 0.056 -> 0.044 | Helped |
| Use TimeGrad instead of Transformer-MAF | Multivariate probabilistic forecasting · Taxi | CRPS_sum: 0.179 -> 0.114 | Helped |
| Use TimeGrad instead of Transformer-MAF | Multivariate probabilistic forecasting · Wikipedia | CRPS_sum: 0.063 -> 0.0485 | Helped |
| Use TimeGrad instead of Transformer-MAF | Multivariate probabilistic forecasting · Exchange | CRPS_sum: 0.005 -> 0.006 | Hurt |
| Reduce the diffusion process length to approximately 10 steps (ablation) | TimeGrad · Electricity | CRPS_sum | No clear effect |
| Increase diffusion steps beyond approximately 100 (ablation) | TimeGrad · Electricity | CRPS_sum | No clear effect |
| Normalize each entity by its context-window mean | TimeGrad · Multivariate time series forecasting | CRPS_sum | Helped |
Closest papers by meaning
- DiffPTS: Rethinking Diffusion ELBO for Probabilistic Time Series Forecasting (NeurIPS 2026 · similarity 0.92)
- Non-stationary Diffusion For Probabilistic Time Series Forecasting (ICML 2025 · similarity 0.92)
- Latent Diffusion Transformer for Probabilistic Time Series Forecasting (AAAI 2024 · similarity 0.91)
- Probabilistic Time Series Modeling with Decomposable Denoising Diffusion Model (ICML 2024 · similarity 0.90)
- Auto-Regressive Moving Diffusion Models for Time Series Forecasting (AAAI 2025 · similarity 0.89)
- Multivariate Probabilistic Time Series Forecasting with Correlated Errors (NeurIPS 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.