> ML_ALGORITHM // NEURAL-BASIS-EXPANSION-N-BEATS_v1.0
N-BEATS (Neural Basis Expansion Analysis)
Deep neural architecture based on backward and forward residual links and hierarchical basis expansion that beat pure statistical methods on the M4 benchmark.
Pure Neural Forecasting Architecturetime-series-forecastinghigh-intrinsiclarge (>100k)
Back to All AlgorithmsComputational Complexity
Training Complexity:O(epochs * batch_size * blocks * MLP)
Inference Complexity:O(blocks * MLP)
Hardware Profile
CPU Friendly:Yes
Requires GPU:Yes
Memory Footprint:moderate
Interpretability & Data
Interpretability Tier:high-intrinsic
Training Data Needs:large (>100k)
Interpretability Assessment
Interpretable configuration restricts output expansion coefficients to explicit monotonic polynomial trend and periodic harmonics.
Suitable Tasks & Supported Modalities
Suitable Tasks:
univariate time series forecastingzero shot ts forecasting
Supported Modalities:
time-series
Implementing Libraries
Foundational Literature
Common Pitfalls & Warnings
- Generic basis configuration sacrifices clear interpretability for marginal accuracy gains