AI Paper Atlas

Research map / Time series / Long-term forecasting

Frequency-domain MLPs are More Effective Learners in Time Series Forecasting

Kun Yi, Qi Zhang, Wei Fan, Shoujin Wang, Pengyang Wang, Hui He, Ning An, Defu Lian, Longbing Cao, Zhendong Niu

NeurIPS 2023 · Time series › Long-term forecasting · approach: Transformers for long-term forecasting · 111 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
remove frequency channel learner (ablation)FreTS · Electricity short-termMAE: 0.050 -> 0.058Hurt
remove frequency temporal learner (ablation)FreTS · Weather 96/336MAE: 0.046 -> 0.051Hurt
add frequency channel learner (ablation)FreTS temporal-only (FreTL) · ETTm1/ETTh1 long horizonsMAE: 0.069 -> 0.071Hurt
add frequency temporal learner (ablation)FreCL · Electricity short-termMAE: 0.054 -> 0.050Improved
replace MLP with FreMLPDLinear · ExchangeMAE: 0.095 -> 0.086Improved
replace MLP with FreMLPDLinear · WeatherMAE: 4.9% in MAEImproved
replace MLP with FreMLPNLinear · ExchangeMAE: 0.115 -> 0.109Improved
FreTS vs best Transformer baselinesFreTS · long-term benchmarksMAE: more than 20%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy