Sometimes a machine learning algorithm or other form of AI may be used as a surrogate expert, to make parameter choices about statistical technigues or neural network architectures; choices that would otherwise be made by a statistical or AI expert. For example, in time series analysis one needs to decide how many time steps to use in windowing methods such as ARMA.
Used in Chap. 14: page 217
Used in glossary entries: ARMA (Auto Regressive Moving Average), time series analysis, windowing
Using ML to choose parameters for statistical processing