GatedLinear: Adaptive Routing of Complementary Linear Bases for Time Series Forecasting
Time series forecasting requires models to capture diverse, often mutually exclusive, temporal dynamics, from smooth trend continuation to nonstationary drift and strict phase-aligned recurrence. While recent deep learning models have improved accuracy, they typically force these diverse patterns through a single computational backbone governed by fixed algorithmic inductive biases (e.g., self-attention or spectral filtering). This single-mechanism approach often struggles with the profound heterogeneity of real-world series, where different variables and forecast horizons necessitate fundamen
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 53%aeon-toolkit/aeon →
- PossiblePossibly related (embedding) · 48%Nixtla/neuralforecast →
- PossiblePossibly related (embedding) · 47%Northern America Deep Learning in Machine Vision - Market Analysis, Forecast, Size, Trends and Insights - IndexBox →
- PossiblePossibly related (embedding) · 46%sktime/sktime →
- LinkedLinked via arxiv author · 85%Qitai Tan →
“GatedLinear: Adaptive Routing of Complementary Linear Bases for Time Series Forecasting”
- LinkedLinked via arxiv author · 85%Ruiwen Gu →
“GatedLinear: Adaptive Routing of Complementary Linear Bases for Time Series Forecasting”
- LinkedLinked via arxiv author · 85%Yilin Su →
“GatedLinear: Adaptive Routing of Complementary Linear Bases for Time Series Forecasting”
- LinkedLinked via arxiv author · 85%Mo Li →
“GatedLinear: Adaptive Routing of Complementary Linear Bases for Time Series Forecasting”