Research map / Time series / Causal and generative time series
Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models
Vincent LE GUEN, Nicolas THOME
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 |
|---|---|---|---|
| train with DILATE instead of MSE | Seq2Seq · Synthetic | TDI: 17.2 -> 14.8 | Improved |
| train with DILATE instead of MSE | Seq2Seq · ECG5000 | DTW: 17.8 -> 16.1 | Improved |
| train with DILATE instead of MSE | Seq2Seq · ECG5000 | MSE: 21.2 -> 30.3 | Hurt |
| train with DILATE instead of DTW_gamma | Seq2Seq · ECG5000 | TDI: 27.2 -> 6.59 | Improved |
| train with DILATE instead of DTW_gamma | MLP/Seq2Seq · all datasets | DTW | No clear effect |
| disentangled shape/time terms instead of weighted-DTW tangled variant (ablation) | DILATE Seq2Seq · Synthetic | TDI: 17.8 -> 14.8 | Improved |
| custom backward pass instead of autodiff | DILATE · prediction length 100 | speedup: x 35 | Improved |
| alpha -> 0 (temporal term only) (ablation) | DILATE · Synthetic | MSE and DTW | Hurt |
| Seq2Seq DILATE vs TT-RNN trained with MSE | forecasting · Traffic | MSE: 0.837 vs 1.00 | Hurt |
Closest papers by meaning
- Loss Shaping Constraints for Long-Term Time Series Forecasting (ICML 2024 · similarity 0.87)
- SynTSBench: Rethinking Temporal Pattern Learning in Deep Learning Models for Time Series (NeurIPS 2025 · similarity 0.86)
- DDN: Dual-domain Dynamic Normalization for Non-stationary Time Series Forecasting (NeurIPS 2024 · similarity 0.85)
- Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts (ICLR 2025 · similarity 0.85)
- LETS Forecast: Learning Embedology for Time Series Forecasting (ICML 2025 · similarity 0.84)
- DBLoss: Decomposition-based Loss Function for Time Series Forecasting (NeurIPS 2025 · similarity 0.84)
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.
