Research map / Time series / Causal and generative time series
Koopa: Learning Non-stationary Time Series Dynamics with Koopman Predictors
Yong Liu, Chenyu Li, Jianmin Wang, Mingsheng Long
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 |
|---|---|---|---|
| add time-variant KP to time-invariant KP only (ablation) | Koopa · ECL | MSE: 0.148 -> 0.146 | Improved |
| add time-variant KP to time-invariant KP only (ablation) | Koopa · ILI | MSE: 2.146 -> 1.734 | Improved |
| add time-invariant KP to time-variant KPs only (ablation) | Koopa · ECL | MSE: 1.547 -> 0.146 | Improved |
| add time-invariant KP to time-variant KPs only (ablation) | Koopa · Traffic | MSE: 0.947 -> 0.419 | Improved |
| use Fourier Filter instead of high-low pass truncated filter (ablation) | Koopa · Traffic | MSE: 0.536 -> 0.419 | Improved |
| use Fourier Filter instead of truncated filter (ablation) | Koopa · ILI | MSE: 1.988 -> 1.734 | Improved |
| use time-invariant then time-variant order instead of switched branches (ablation) | Koopa · ECL | MSE: 0.696 -> 0.146 | Improved |
| remove reconstruction loss (Koopa predictor vs Koopman Autoencoder) (ablation) | Koopa · ILI | MSE: 2.189 -> 1.734 | Improved |
| remove reconstruction loss (Koopa predictor vs Koopman Autoencoder) (ablation) | Koopa · Exchange | MSE: 0.129 -> 0.110 | Improved |
| add operator adaptation when scaling up forecast horizon (ablation) | Koopa · Exchange | MSE: 0.214 -> 0.172 | Improved |
| add operator adaptation when scaling up horizon (ablation) | Koopa · Traffic | MSE: 0.709 -> 0.699 | Improved |
| use Koopa instead of PatchTST | PatchTST · ECL H=192 | MSE: 0.147 -> 0.156 | Hurt |
Later papers that compared with this method
7 later accepted papers report Koopa as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.
- KUMA: A Novel Framework with Koopman Separation and Efficient Multilevel Extraction in Time Series Forecasting (ICML 2026 · better on 11 of 12 shared results)
- Ellipsoidal Time Series Forecasting (ICML 2026 · better on 2 of 3 shared results)
- SKOLR: Structured Koopman Operator Linear RNN for Time-Series Forecasting (ICML 2025 · better on 11 of 11 shared results)
- KoNODE: Koopman-Driven Neural Ordinary Differential Equations with Evolving Parameters for Time Series Analysis (ICML 2025 · better on 2 of 3 shared results)
- LETS Forecast: Learning Embedology for Time Series Forecasting (ICML 2025 · better on 1 of 3 shared results)
- Attractor Memory for Long-Term Time Series Forecasting: A Chaos Perspective (NeurIPS 2024 · better on 2 of 2 shared results)
- AutoTimes: Autoregressive Time Series Forecasters via Large Language Models (NeurIPS 2024 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- TimesNet — TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis (ICLR 2023) · better on 6 of 6 shared results
- N-HiTS — NHITS: Neural Hierarchical Interpolation for Time Series Forecasting (AAAI 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SKOLR: Structured Koopman Operator Linear RNN for Time-Series Forecasting (ICML 2025 · similarity 0.90)
- KoNODE: Koopman-Driven Neural Ordinary Differential Equations with Evolving Parameters for Time Series Analysis (ICML 2025 · similarity 0.89)
- $K^2$VAE: A Koopman-Kalman Enhanced Variational AutoEncoder for Probabilistic Time Series Forecasting (ICML 2025 · similarity 0.88)
- Deep-Koopman-KANDy: Dictionary Discovery for Deep-Koopman Operators with Kolmogorov-Arnold Networks for Dynamics (NeurIPS 2026 · similarity 0.88)
- Shifting Time: Time-series Forecasting with Khatri-Rao Neural Operators (ICML 2025 · similarity 0.87)
- Course Correcting Koopman Representations (ICLR 2024 · similarity 0.86)
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.
