What is the difference of a feature vector and a 'vector-valued observation' as described here?

The term 'vector-valued' is used in the following context:

"Most state-of-the-art [Automatic Speech Recognition] systems use vector valued observations, which are modeled with Gaussian mixture emission densities."


Saying that something is vector-valued is simply another way of saying that something is a vector, e.g. in regression model

$$ \mathbf{y} = \mathbf{X}\boldsymbol{\beta} + \boldsymbol{\varepsilon} $$

$\mathbf{y}, \boldsymbol{\beta}, \boldsymbol{\varepsilon}$ are all vectors and $\mathbf{X}$ is a matrix, so $\boldsymbol{\beta} = (\beta_0, \beta_1, \dots, \beta_k)$ is vector-valued parameter.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.