Chapter 11 Probability cheatsheet

11.1 Gaussian distribution

Definition 11.1 (Multivariate Gaussian distribution) N(x;μ,Σ)=1(2π)n/2|Σ|1/2exp(12(xμ)Σ1(xμ))

where x,μRn, ΣRn×n

Linear combination Ax of Gaussian x is Gaussian with:

  • mean: Aμ
  • variance: AΣA

The conditional expectation of Gaussian with respect to Ax=y is Gaussian with

  • conditional mean: μ+ΣA(AΣA)1(yAμ)
  • conditional variance: ΣΣA(AΣA)1AΣ

As a result, one obtains the following general formula:

p(Bx|Ax=y)=N(Bx;Bμ+BΣA(AΣA)1(yAμ),BΣBBΣA(AΣA)1AΣB)

Note also that E(exp(λX))=exp(λμ+λ2σ22).