Research map / Time series / Causal and generative time series
Probabilistic Forecasting: A Level-Set Approach
Hilaf Hasson, Bernie Wang, Tim Januschowski, Jan Gasthaus
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 XLSF instead of conformalized predictions | tabular quantile prediction · facebook1 | P05 loss: 0.471 -> 0.103 | Improved |
| use XLSF instead of conformalized predictions | tabular quantile prediction · facebook1 | P95 loss: 0.428 -> 0.288 | Improved |
| use XLSF instead of conformalized predictions | tabular quantile prediction · facebook1 | interval accuracy: 94.31% vs 89.98% | Hurt |
| use XLSF instead of QRF | tabular quantile prediction · facebook2 | training time (s): 103.61 -> 14.790 | Improved |
| use XLSF instead of QRF | tabular quantile prediction · facebook1 | P95 loss: 0.317 -> 0.288 | Improved |
| use XLSF instead of QRF | tabular quantile prediction · facebook1 | P05 loss: 0.094 vs 0.103 | Hurt |
| wrap M5 accuracy winner with LSF | M5 winning point forecaster · M5 uncertainty bottom level | mean_wQL: 0.35551 | No clear effect |
| TTSW (XLSF) instead of DeepAR | TTSW · electricity | mean wQL: 0.0592 -> 0.0499 | Improved |
| TTSW (XLSF) instead of DeepAR | TTSW · traffic | mean wQL: 0.0913 -> 0.1212 | Hurt |
| TTSW (XLSF) instead of CNN-QR | TTSW · parts | mean wQL: 0.7566 -> 0.9709 | Hurt |
| TTSW (XLSF) instead of CNN-QR | TTSW · m4_daily | mean wQL: 0.0154 -> 0.0148 | Improved |
| TTSW with quantile regression instead of XLSF | TTSW · electricity | mean wQL: 0.0499 -> 0.1478 | Hurt |
Closest papers by meaning
- ProbTS: Benchmarking Point and Distributional Forecasting across Diverse Prediction Horizons (NeurIPS 2024 · similarity 0.87)
- HDT: Hierarchical Discrete Transformer for Multivariate Time Series Forecasting (AAAI 2025 · similarity 0.84)
- TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting (ICML 2025 · similarity 0.84)
- Cherry-Picking in Time Series Forecasting: How to Select Datasets to Make Your Model Shine (AAAI 2025 · similarity 0.83)
- Multivariate Probabilistic Time Series Forecasting with Correlated Errors (NeurIPS 2024 · similarity 0.83)
- TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Data (AAAI 2025 · similarity 0.83)
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.
