One of the underlying assumptions of PCA is linearity, such that the feature space is a linear transformation of the original data.
Through a linear transformation of the concentration vector a new model is set up.
For reasons of computational and conceptual simplicity, the representation is often sought as a linear transformation of the original data.
In representational measurement theory, the current theory of all measurements, calibration and sampling processes are assumed to be a linear transformation of the coordinate system, of no effect.
In addition, MLLT finds a linear transformation of features to reduce state-conditional feature correlations.
Finally, a linear transformation of atmospheric waves into magnetospheric waves is possible in principle.
Then (1.3) provides a linear transformation of the observation data, into the best estimate for every.
LIN simply performs a linear transformation of context expended input features.
It is a linear transformation of original data, the result value mapping between [0, 1]. .
Without quantization, the embedding calculation is a linear transformation of the input profile.
In such solutions, the noise-free speech is estimated by a linear transformation of the observation vector.
