A Comprehensive Survey of Deep Learning for Multivariate Time Series Forecasting: A Channel Strategy Perspective

January 19, 2026 · View on GitHub

PRs Welcome Stars

image-20250212190349685

  • A channel strategy refers to the approach employed to process, integrate, or utilize information from multiple input channels.
  • Channel Independence (CI): The CI strategy treats each channel independently, without considering any potential interactions or correlations among channels. Each channel is processed as a separate input, and no shared information or dependencies are utilized.
  • Channel Dependence (CD): The CD strategy assumes that all channels in a multivariate time series are inherently correlated and interdependent, treating them as a unified entity during the forecasting process.
  • Channel Partiality (CP): The CP strategy strikes a balance between CI and CD, allowing each channel to retain a degree of independence while simultaneously interacting with other related channels. This approach emphasizes a hybrid state where channels selectively interact and exhibit partial correlations.

This repository contains a list of papers and their corresponding codebases, categorized by CI, CD, and CP.

StrategyMechanismAsym.Lag.Pol.Gw.Dyn.Ms.MethodParadigmVenueYearCode
CI-------PatchTSTSpecificICLR2023Link
CI-------CycleNetSpecificNIPS2024Link
CI-------DLinearSpecificAAAI2023Link
CI-------One-Fits-AllFoundationICML2024Link
CI-------TimerFoundationNIPS2023Link
CI-------S2IP-LLMFoundationICML2024Link
CI-------TEMPOFoundationICML2024Link
CI-------LLM4TSFoundationNIPS2023Link
CI-------Time-LLMFoundationICLR2024Link
CI-------RevINPluginICLR2021Link
CDCNN-based✓-----InformerSpecificAAAI2021Link
CDCNN-based✓-----AutoformerSpecificNIPS2021Link
CDCNN-based✓-----FEDformerSpecificICML2022Link
CDCNN-based✓-----TimesNetSpecificICLR2023Link
CDMLP-based✓-----TSMixerSpecificKDD2023Link
CDMLP-based✓-----TTMFoundationNIPS2024Link
CDTransformer-based✓---✓-iTransformerSpecificICLR2024Link
CDTransformer-based✓---✓-CrossformerSpecificICLR2023Link
CDTransformer-based✓✓----VCformerSpecificIJCAI2024Link
CDTransformer-based✓✓--✓-MOIRAIFoundationICML2024Link
CDTransformer-based✓---✓-UniTSFoundationNIPS2024Link
CDTransformer-based✓---✓-TQNetSpecificICML2025Link
CDGNN-based----✓-GTSSpecificICLR2021Link
CDGNN-based✓----✓MSGNetSpecificAAAI2024Link
CDGNN-based-✓----FourierGNNSpecificNIPS2023Link
CDGNN-based-✓----FC-STGNNSpecificAAAI2024Link
CDGNN-based✓---✓-TPGNNSpecificNIPS2022Link
CDGNN-based✓---✓✓ESGSpecificKDD2022Link
CDGNN-based✓---✓✓EnhanceNetPluginICDE2021Link
CDOthers✓✓----SOFTSSpecificNIPS2024Link
CDOthers✓--✓--C-LoRAPluginCIKM2024Link
CPCNN-based✓-----ModernTCNSpecificICLR2024Link
CPTransformer-based✓--✓--DUETSpecificKDD2025Link
CPTransformer-based✓---✓-MCformerSpecificIITJ*2024-
CPTransformer-based✓--✓✓-DGCformerSpecificarXiv2024-
CPTransformer-based----✓-CMPluginNIPS2024-
CPGNN-based✓---✓-MTGNNSpecificKDD2020Link
CPGNN-based✓--✓--CrossGNNSpecificNIPS2023Link
CPGNN-based✓---✓-WaveForMSpecificAAAI2023Link
CPGNN-based----✓-MTSF-DGSpecificVLDB2023Link
CPGNN-based✓--✓✓✓ReMoSpecificIJCAI2023-
CPGNN-based✓---✓✓Ada-MSHyperSpecificNIPS2024Link
CPGNN-based-✓-✓✓-TimeFilterSpecificICML2025Link
CPOthers✓✓--✓-LIFTPluginICLR2024Link
CPOthers✓--✓--CCMPluginNIPS2024Link