mlsanityData to 7 Oct 2026

FeDaL: Federated Dataset Learning for General Time Series Foundation Models

arXiv 2508.04045 ↗First posted 6 Aug 2025cs.LGcited by 10

Not on the map. Why?

Abstract (arXiv)

Dataset-level heterogeneity introduces significant domain biases that fundamentally degrade generalization on general Time Series Foundation Models (TSFMs), yet this challenge remains underexplored. This paper rethinks the from-scratch training of TSFMs using the paradigm of federated learning. We propose a novel Federated Dataset Learning (FeDaL) approach to tackle heterogeneous time series by learning dataset-agnostic temporal representations. Specifically, the distributed architecture of federated learning is a nature solution to decompose heterogeneous TS datasets into shared generalized knowledge and preserved personalized knowledge. Moreover, based on the TSFM architecture, FeDaL explicitly mitigates both local and global biases by adding two complementary mechanisms: Domain Bias Elimination (DBE) and Global Bias Elimination (GBE). FeDaL`s cross-dataset generalization has been extensively evaluated in real-world datasets spanning eight tasks (including various regression and classification), against 54 baselines. We further analyze federated scaling behavior, showing how data volume, client count, and join rate affect model performance under decentralization. Our code is publicly available at https://github.com/shengchaochen82/FeDaL

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 64 of its 79 references that are in the map. First posted on arXiv 6 Aug 2025.

Builds on

64 of 79 references.

References · 64

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  4. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  5. 2020Scaling Laws for Neural Language Modelscited by 6,411
  6. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  7. 2018Federated Optimization in Heterogeneous Networkscited by 2,971
  8. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,424
  9. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  10. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  11. 2020Reformer: The Efficient Transformercited by 1,658
  12. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,654
38 more
  1. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
  2. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,200
  3. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,030
  4. 2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 869
  5. 2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 744
  6. 2024Chronos: Learning the Language of Time Seriescited by 712
  7. 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 680
  8. 2020Ditto: Fair and Robust Federated Learning Through Personalizationcited by 665
  9. 2023A decoder-only foundation model for time-series forecastingcited by 650
  10. 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 615
  11. 2019Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecastingcited by 606
  12. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
  13. 2021FedBN: Federated Learning on Non-IID Features via Local Batch Normalizationcited by 565
  14. 2021Federated Learning Based on Dynamic Regularizationcited by 555
  15. 2021Exploiting Shared Representations for Personalized Federated Learningcited by 537
  16. 2019Federated Learning with Personalization Layerscited by 533
  17. 2021TS2Vec: Towards Universal Representation of Time Seriescited by 463
  18. 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 453
  19. 2018The UCR Time Series Archivecited by 421
  20. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 387
  21. 2022Non-stationary Transformers: Exploring the Stationarity in Time Series Forecastingcited by 385
  22. 2021Time-Series Representation Learning via Temporal and Contextual Contrastingcited by 342
  23. 2019ROCKET: Exceptionally fast and accurate time series classification using random convolutional kernelscited by 336
  24. 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 333
  25. 2018The UEA multivariate time series classification archive, 2018cited by 330
  26. 2024Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Expertscited by 280
  27. 2021GRAD-MATCH: Gradient Matching based Data Subset Selection for Efficient Deep Model Trainingcited by 269
  28. 2019Unsupervised Scalable Representation Learning for Multivariate Time Seriescited by 267
  29. 2017Soft-DTW: a Differentiable Loss Function for Time-Seriescited by 265
  30. 2021Unsupervised Representation Learning for Time Series with Temporal Neighborhood Codingcited by 223
  31. 2020Personalized Federated Learning with First Order Model Optimizationcited by 201
  32. 2022Less Is More: Fast Multivariate Time Series Forecasting with Light Sampling-oriented MLP Structurescited by 158
  33. 2023TimeGPT-1cited by 149
  34. 2022N-HiTS: Neural Hierarchical Interpolation for Time Series Forecastingcited by 146
  35. 2022ETSformer: Exponential Smoothing Transformers for Time-series Forecastingcited by 146
  36. 2023UniTime: A Language-Empowered Unified Model for Cross-Domain Time Series Forecastingcited by 139
  37. 2020Lower Bounds and Optimal Algorithms for Personalized Federated Learningcited by 122
  38. 2024TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysiscited by 113

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 10.

10 papers.

  1. 2026Bi-level Heterogeneous Learning for Time Series Foundation Models: A Federated Learning Approachcited by 2
  2. 2026OmniFM: Toward Modality-Robust and Task-Agnostic Federated Learning for Heterogeneous Medical Imagingcited by 1
  3. 2026Spectral-Aware Text-to-Time Series Generation with Billion-Scale Multimodal Meteorological Datacited by 1
  4. 2026On the Tradeoffs of On-Device Generative Models in Federated Predictive Maintenance Systemscited by 1
  5. 2026PiXTime: A Model for Federated Time Series Forecasting with Heterogeneous Data across Nodescited by 0
  6. 2026Federated Weather Modeling on Sensor Datacited by 0
  7. 2026FM2: Unified Federated Foundation Models for Heterogeneous Multimodal Medical Imagingcited by 0
  8. 2026Lethe: How Hard Is It to Forget? A Benchmark for Federated Unlearning in Medical Imagingcited by 0
  9. 2026Copy What Is Seen, Generate What Is Not: Training-Free Anomaly-Aware Video Restorationcited by 0
  10. 2026CORD: Learning Reusable Degradation Representations Across Heterogeneous Physical Systemscited by 0