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

arXiv: https://arxiv.org/abs/2002.02887 · HTML: https://mlsanity.com/paper/2002.02887/

## Key facts

- Cited by 70 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Feb 2020.

## Built on by (most cited first, 30 of 70)

- [A decoder-only foundation model for time-series forecasting](https://mlsanity.com/paper/2310.10688/) (2023)
- [One Fits All:Power General Time Series Analysis by Pretrained LM](https://mlsanity.com/paper/2302.11939/) (2023)
- [MOMENT: A Family of Open Time-series Foundation Models](https://mlsanity.com/paper/2402.03885/) (2024)
- [Chronos-2: From Univariate to Universal Forecasting](https://mlsanity.com/paper/2510.15821/) (2025)
- [TimeGPT-1](https://mlsanity.com/paper/2310.03589/) (2023)
- [AdaRNN: Adaptive Learning and Forecasting of Time Series](https://mlsanity.com/paper/2108.04443/) (2021)
- [ForecastPFN: Synthetically-Trained Zero-Shot Forecasting](https://mlsanity.com/paper/2311.01933/) (2023)
- [Darts: User-Friendly Modern Machine Learning for Time Series](https://mlsanity.com/paper/2110.03224/) (2021)
- [Domain Adaptation for Time Series Forecasting via Attention Sharing](https://mlsanity.com/paper/2102.06828/) (2021)
- [PSA-GAN: Progressive Self Attention GANs for Synthetic Time Series](https://mlsanity.com/paper/2108.00981/) (2021)
- [A Survey on Time-Series Pre-Trained Models](https://mlsanity.com/paper/2305.10716/) (2023)
- [Long-Range Transformers for Dynamic Spatiotemporal Forecasting](https://mlsanity.com/paper/2109.12218/) (2021)
- [Retrieval Augmented Time Series Forecasting](https://mlsanity.com/paper/2411.08249/) (2024)
- [Koopman Neural Forecaster for Time Series with Temporal Distribution Shifts](https://mlsanity.com/paper/2210.03675/) (2022)
- [TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Data](https://mlsanity.com/paper/2502.16294/) (2025)
- [Zero-shot forecasting of chaotic systems](https://mlsanity.com/paper/2409.15771/) (2024)
- [Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain](https://mlsanity.com/paper/2310.05063/) (2023)
- [Towards Urban General Intelligence: A Review and Outlook of Urban Foundation Models](https://mlsanity.com/paper/2402.01749/) (2024)
- [Scaling-laws for Large Time-series Models](https://mlsanity.com/paper/2405.13867/) (2024)
- [Context parroting: A simple but tough-to-beat baseline for foundation models in scientific machine learning](https://mlsanity.com/paper/2505.11349/) (2025)
- [Mamba4Cast: Efficient Zero-Shot Time Series Forecasting with State Space Models](https://mlsanity.com/paper/2410.09385/) (2024)
- [Out-of-Distribution Generalization in Time Series: A Survey](https://mlsanity.com/paper/2503.13868/) (2025)
- [Quadratic Direct Forecast for Training Multi-Step Time-Series Forecast Models](https://mlsanity.com/paper/2511.00053/) (2025)
- [TimeCopilot](https://mlsanity.com/paper/2509.00616/) (2025)
- [LaT-PFN: A Joint Embedding Predictive Architecture for In-context Time-series Forecasting](https://mlsanity.com/paper/2405.10093/) (2024)
- [Multimodal Meta-Learning for Time Series Regression](https://mlsanity.com/paper/2108.02842/) (2021)
- [Few-Shot Forecasting of Time-Series with Heterogeneous Channels](https://mlsanity.com/paper/2204.03456/) (2022)
- [FusionSF: Fuse Heterogeneous Modalities in a Vector Quantized Framework for Robust Solar Power Forecasting](https://mlsanity.com/paper/2402.05823/) (2024)
- [TRACE: Time SeRies PArameter EffiCient FinE-tuning](https://mlsanity.com/paper/2503.16991/) (2025)
- [Meta-Forecasting by combining Global Deep Representations with Local Adaptation](https://mlsanity.com/paper/2111.03418/) (2021)

## 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.