Expectation of Gaussian Random Vectors

56 Views Asked by At

Let $X\sim N(0,I_n)$ be a standard normal random vector. Also, let $a$ and $b$ be two arbitrary vectors in $\mathbb{R}^n$. Moreover define the ReLU function as follows: ReLU($x)=x$ if $x>0$ otherwise ReLU$(x)=0$.

I want to compute the following expectation:

$$\mathbb{E}[\text{ReLU}(a^TX)\text{ReLU}(b^TX)].$$

I know that $a^TX\sim N(0,||a||^2)$ and $b^TX\sim N(0,||b||^2)$. I don't know how to find joint distribution of $a^TX$ and $b^TX$.