Pith. sign in

REVIEW 3 cited by

FreDF: Learning to Forecast in the Frequency Domain

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.02399 v2 pith:NT76U5NZ submitted 2024-02-04 cs.LG cs.AIstat.APstat.ML

classification cs.LGcs.AIstat.APstat.ML
keywords autocorrelationforecastfredflabellearningdatadirectdomain
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Time series modeling presents unique challenges due to autocorrelation in both historical data and future sequences. While current research predominantly addresses autocorrelation within historical data, the correlations among future labels are often overlooked. Specifically, modern forecasting models primarily adhere to the Direct Forecast (DF) paradigm, generating multi-step forecasts independently and disregarding label autocorrelation over time. In this work, we demonstrate that the learning objective of DF is biased in the presence of label autocorrelation. To address this issue, we propose the Frequency-enhanced Direct Forecast (FreDF), which mitigates label autocorrelation by learning to forecast in the frequency domain, thereby reducing estimation bias. Our experiments show that FreDF significantly outperforms existing state-of-the-art methods and is compatible with a variety of forecast models. Code is available at https://github.com/Master-PLC/FreDF.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Dual-Prototype Disentanglement: A Context-Aware Enhancement Framework for Time Series Forecasting

    cs.LG 2026-01 conditional novelty 6.0 of 10

    A model-agnostic module that retrieves common and rare prototype patterns improves forecasting error on many standard benchmarks, but not on all reported cases.

  2. KARMA: A Multilevel Decomposition Hybrid Mamba Framework for Multivariate Long-Term Time Series Forecasting

    cs.LG 2025-06 conditional novelty 6.0 of 10

    KARMA combines adaptive trend/seasonal decomposition, wavelet frequency decomposition, and Mamba blocks to forecast long multivariate time series, reporting state-of-the-art accuracy on several benchmarks.

  3. TimeCF: A TimeMixer-Based Model with adaptive Convolution and Sharpness-Aware Minimization Frequency Domain Loss for long-term time seris forecasting

    cs.LG 2025-05 conditional novelty 4.0 of 10

    TimeCF, a TimeMixer-based forecaster with adaptive convolutions and a sharpness-aware frequency-domain loss, reports modest MSE/MAE gains over eight baselines on six long-term forecasting datasets.

Pith tools