Research map / Time series / Causal and generative time series
An Extreme-Adaptive Time Series Prediction Model Based on Probability-Enhanced LSTM Neural Networks
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 NEC+ instead of the strongest baseline | 3-day reservoir water-level forecasting · Reservoirs 4001–4010 | RMSE | Helped |
| Use N-BEATS instead of NEC+ | 3-day reservoir water-level forecasting · Reservoir 4011 | RMSE | Hurt |
| Add the GMM indicator to NEC (ablation) | NEC · Nine reservoirs | RMSE: 18% average RMSE improvement | Helped |
| Add the GMM indicator to LSTM (ablation) | LSTM · Nine reservoirs | RMSE: 4% average RMSE improvement | Mixed |
| Add watershed rain gauge variables (ablation) | NEC+G · Reservoir 4005 | RMSE: 5114.49 -> 4352.74 | Helped |
| Add watershed rain gauge variables (ablation) | NEC+G · Reservoir 4007 | RMSE: 2924.30 -> 2092.73 | Helped |
| Add watershed rain gauge variables (ablation) | NEC+G · Reservoir 4010 | RMSE: 2385.77 -> 2275.48 | Helped |
| Add watershed rain gauge variables (ablation) | LSTM · Reservoirs 4005, 4007, and 4010 | RMSE | Hurt |
| Increase oversampling of extreme-containing samples (ablation) | LSTM and LSTM+G · Sensor 4005 | Total RMSE: OS=0.04 gave 7919.8 (L+G) and 7769.6 (L); higher oversampling worsened total RMSE | Mixed |
| Increase the loss alpha parameter (ablation) | NEC+ classifier · Sensor 4004 | extreme classifications | Helped |
| Set loss alpha too large (ablation) | NEC+ classifier · Sensor 4004 | prediction performance | Mixed |
Closest papers by meaning
- Learning from Polar Representation: An Extreme-Adaptive Model for Long-Term Time Series Forecasting (AAAI 2024 · similarity 0.88)
- Unlocking the Power of LSTM for Long Term Time Series Forecasting (AAAI 2025 · similarity 0.82)
- E4GEN: Event-level Explainable Extreme-Enhanced Time-series Generation (NeurIPS 2026 · similarity 0.82)
- LETS Forecast: Learning Embedology for Time Series Forecasting (ICML 2025 · similarity 0.82)
- FIDE: Frequency-Inflated Conditional Diffusion Model for Extreme-Aware Time Series Generation (NeurIPS 2024 · similarity 0.82)
- xPatch: Dual-Stream Time Series Forecasting with Exponential Seasonal-Trend Decomposition (AAAI 2025 · similarity 0.82)
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.