Table 1 The models used in NAMs and the number of parameters of each model based on the number of features (s), hidden neurons (H), and input and output sequence length n
From: Short-term prediction of celestial pole offsets with interpretable machine learning
Model | Number of parameters |
|---|---|
\(\mu _j(x_k)\) | \(4H^2+5nH+4H+n\) |
\(\sigma _j^2(x_k)\) | \(4H^2+5nH+4H+n\) |
\(\mu _j\) | \(s(4H^2+5nH+4H+n)\) |
\(\sigma _j^2\) | \(s(4H^2+5nH+4H+n)\) |
NAMs | \(2s(4H^2+5nH+4H+n)\) |