mlsanityData to 7 Oct 2026

FAME: Forecastability-Aware Mixture of Experts for Heterogeneous Time Series Forecasting

arXiv 2606.08896 ↗First posted 8 Jun 2026cs.AIcited by 0

On the map: Mixture-of-experts → MoE for time series and graphs

Abstract (arXiv)

Large-scale retail and industrial forecasting systems contain many heterogeneous time series whose lifecycle, sparsity, volatility, seasonality, spectral patterns, and contextual sensitivity differ substantially. A single forecasting model rarely performs well across all regimes, while dense ensembles increase inference cost and provide limited insight into expert suitability. This paper studies forecastability-aware expert routing: learning how data characteristics determine the suitability of forecasting experts. We propose method, a sparse mixture-of-experts framework that represents each series with a multidimensional forecastability fingerprint, mines expert-suitability targets from validation performance, and trains a cost-aware sparse router to activate a small budgeted set of experts for each series. Using a production-scale vending-machine sales dataset from Shandong New Beiyang (SNBC), where the forecasting component has been integrated into the replenishment-planning pipeline, together with public retail benchmarks, we show that expert suitability varies systematically across data regimes. On the industrial dataset with 5,000+ machines and 60M+ transactions, method Top-2 reduces MSE by 12.4% over the strongest single expert, LightGBM, while executing 1.92 experts per series on average. The deployed component produces demand forecasts, while inventory-oriented gains are estimated by an offline replay simulator under a fixed replenishment policy rather than by online intervention. The framework turns heterogeneous sales forecasting from heuristic model selection into data mining of forecastability patterns and expert specialization. Code is available at https://github.com/hit636/FAME

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in MoE for time series and graphs, one of 12 directions in Mixture-of-experts. Builds on 17 of its 32 references that are in the map. First posted on arXiv 8 Jun 2026.

Builds on

17 of 32 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025GateTS: Versatile and Efficient Forecasting via Attention-Inspired routed Mixture-of-Experts24 Aug
  2. 2024Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts24 Sep

Other references · 15

  1. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  2. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  3. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
  4. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,030
  5. 2024Chronos: Learning the Language of Time Seriescited by 712
  6. 2023A decoder-only foundation model for time-series forecastingcited by 650
  7. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
  8. 2020AutoGluon-Tabular: Robust and Accurate AutoML for Structured Datacited by 368
  9. 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 333
  10. 2018Meta-Learning: A Surveycited by 254
  11. 2024TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methodscited by 242
  12. 2024Is Mamba Effective for Time Series Forecasting?cited by 110
3 more
  1. 2024TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecastingcited by 63
  2. 2023AutoGluon-TimeSeries: AutoML for Probabilistic Time Series Forecastingcited by 36
  3. 2024Mamba4Cast: Efficient Zero-Shot Time Series Forecasting with State Space Modelscited by 17

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting
  2. 2026Fast Training of Mixture-of-Experts for Time Series Forecasting via Expert Loss Integration
  3. 2026Fuzzy-MoE: Interpretable Regime-Conditioned Expert Routing for Non-Stationary Multivariate Time Series Forecasting
  4. 2025MoGU: Mixture-of-Gaussians with Uncertainty-based Gating for Time Series Forecasting
  5. 2025MoFE-Time: Mixture of Frequency Domain Experts for Time-Series Forecasting Models
  6. 2026Dynamic TMoE: A Drift-Aware Dynamic Mixture of Experts Framework for Non-Stationary Time Series Forecasting
  7. 2025The Forecast Critic: Leveraging Large Language Models for Poor Forecast Identification
  8. 2023Mixture-of-Linear-Experts for Long-term Time Series Forecasting