I am writing this as a mathematician trying to understand fermionic Gaussian states.
Up to global phase, a quantum state can be faithfully represented in terms of a quasi-probability distribution on its phase space by its Wigner function.
In particular, if this quantum state is a (bosonic) Gaussian state, then this restricts to a multivariate Gaussian probability distribution on the phase space. In other words, the state is described by a (positive definite) matrix determining the covariance between the positions and momentums between modes, as well as a vector determining the mean of this probability distribution. The evolution of these states is closed under the action of Gaussian unitaries, which can be regarded as the probabilistic evolution of the phase space.
However, in the literature, there is a notion of fermionic Gaussian states. See the article of Hackl and Bianchi for reference. However, the covariance matrix is given by a skew-symmetric matrix. Is there any way in which these fermionic Gaussian states can be represented by probability distributions in such a way that their evolution by fermionic unitaries corresponds to the probabilistic evolution of the phase space? Or does the name Gaussian in this case only come from the relation to bosonic fermions, and not Gaussian probability theory?