mlsanityData to 13 Oct 2026

Meta-learning framework with applications to zero-shot time-series forecasting

arXiv 2002.02887 ↗First posted 7 Feb 2020cs.LG stat.MLcited by 70

Not on the map. Why?

Abstract (arXiv)

Can meta-learning discover generic ways of processing time series (TS) from a diverse dataset so as to greatly improve generalization on new TS coming from different datasets? This work provides positive evidence to this using a broad meta-learning framework which we show subsumes many existing meta-learning algorithms. Our theoretical analysis suggests that residual connections act as a meta-learning adaptation mechanism, generating a subset of task-specific parameters based on a given TS input, thus gradually expanding the expressive power of the architecture on-the-fly. The same mechanism is shown via linearization analysis to have the interpretation of a sequential update of the final linear layer. Our empirical results on a wide range of data emphasize the importance of the identified meta-learning mechanisms for successful zero-shot univariate forecasting, suggesting that it is viable to train a neural network on a source TS dataset and deploy it on a different target TS dataset without retraining, resulting in performance that is at least as good as that of state-of-practice univariate forecasting models.

Key facts. Cited by 70 papers in the map (data through 13 Oct 2026). First posted on arXiv 7 Feb 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2021: 12; 2022: 4; 2023: 13; 2024: 18; 2025: 17; 2026: 6.

70 papers.

Sort:

By area:

  1. 2023A decoder-only foundation model for time-series forecastingcited by 674
  2. 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 623
  3. 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 468
  4. 2025Chronos-2: From Univariate to Universal Forecastingcited by 177
  5. 2023TimeGPT-1cited by 153
  6. 2021AdaRNN: Adaptive Learning and Forecasting of Time Seriescited by 120
  7. 2023ForecastPFN: Synthetically-Trained Zero-Shot Forecastingcited by 104
  8. 2021Darts: User-Friendly Modern Machine Learning for Time Seriescited by 91
  9. 2021Domain Adaptation for Time Series Forecasting via Attention Sharingcited by 68
  10. 2021PSA-GAN: Progressive Self Attention GANs for Synthetic Time Seriescited by 58
  11. 2023A Survey on Time-Series Pre-Trained Modelscited by 52
  12. 2021Long-Range Transformers for Dynamic Spatiotemporal Forecastingcited by 46
38 more
  1. 2024Retrieval Augmented Time Series Forecastingcited by 44
  2. 2022Koopman Neural Forecaster for Time Series with Temporal Distribution Shiftscited by 35
  3. 2024Zero-shot forecasting of chaotic systemscited by 28
  4. 2025TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Datacited by 27
  5. 2023Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domaincited by 23
  6. 2024Towards Urban General Intelligence: A Review and Outlook of Urban Foundation Modelscited by 23
  7. 2024Scaling-laws for Large Time-series Modelscited by 17
  8. 2024Mamba4Cast: Efficient Zero-Shot Time Series Forecasting with State Space Modelscited by 17
  9. 2025Context parroting: A simple but tough-to-beat baseline for foundation models in scientific machine learningcited by 17
  10. 2025Out-of-Distribution Generalization in Time Series: A Surveycited by 15
  11. 2025Quadratic Direct Forecast for Training Multi-Step Time-Series Forecast Modelscited by 10
  12. 2025TimeCopilotcited by 9
  13. 2024LaT-PFN: A Joint Embedding Predictive Architecture for In-context Time-series Forecastingcited by 8
  14. 2021Multimodal Meta-Learning for Time Series Regressioncited by 7
  15. 2022Few-Shot Forecasting of Time-Series with Heterogeneous Channelscited by 7
  16. 2024FusionSF: Fuse Heterogeneous Modalities in a Vector Quantized Framework for Robust Solar Power Forecastingcited by 6
  17. 2025TRACE: Time SeRies PArameter EffiCient FinE-tuningcited by 6
  18. 2021Meta-Forecasting by combining Global Deep Representations with Local Adaptationcited by 5
  19. 2023Cross-Frequency Time Series Meta-Forecastingcited by 5
  20. 2023Deep Non-Parametric Time Series Forecastercited by 5
  21. 2023Handling Concept Drift in Global Time Series Forecastingcited by 4
  22. 2024Towards Foundation Time Series Model: To Synthesize Or Not To Synthesize?cited by 4
  23. 2025Not All Data are Good Labels: On the Self-supervised Labeling for Time Series Forecastingcited by 4
  24. 2026Zero-shot Forecasting by Simulation Alonecited by 4
  25. 2021Forecasting Market Prices using DL with Data Augmentation and Meta-learning: ARIMA still wins!cited by 3
  26. 2022Gaussian Process Latent Variable Modeling for Few-shot Time Series Forecastingcited by 3
  27. 2024The impact of data set similarity and diversity on transfer learning success in time series forecastingcited by 3
  28. 2024Urban traffic analysis and forecasting through shared Koopman eigenmodescited by 3
  29. 2025SeqFusion: Sequential Fusion of Pre-Trained Models for Zero-Shot Time-Series Forecastingcited by 3
  30. 2025EnergyPatchTST: Multi-scale Time Series Transformers with Uncertainty Estimation for Energy Forecastingcited by 3
  31. 2025One-Embedding-Fits-All: Efficient Zero-Shot Time Series Forecasting by a Model Zoocited by 3
  32. 2026Why Do Time Series Models Need Long Context Windows?cited by 3
  33. 2026Post-Training in Time Series Foundation Models: A Unifying Frameworkcited by 3
  34. 2024Nonstationary Time Series Forecasting via Unknown Distribution Adaptationcited by 2
  35. 2024Learning to learn ecosystems from limited data -- a meta-learning approachcited by 2
  36. 2025Tailored Forecasting from Short Time Series via Meta-learningcited by 2
  37. 2025MetaSTNet: Multimodal Meta-learning for Cellular Traffic Conformal Predictioncited by 2
  38. 2025A More Realistic Evaluation of Cross-Frequency Transfer Learning and Foundation Forecasting Modelscited by 2

50 most cited of 70 All 70 (JSON)