Can we get some Matrix<T> broadcasting?

[francisco-chavez]

I've been studying Neural Networks, and one of the things we do a lot of is Z = (W*X) + b, where (W * X) results in a matrix that's (n, m), but b is only (n, 1). Broadcasting allows us to add b to each column of (W * X), so it basically teats b like an (n, m) matrix where each column is an identical.

I'm able to write code that does this manually, but I don't think the code I wrote is very well optimized, which defeats the purpose of using Math.Net.

Is there a way we can get some kind of broadcasting method added to Matrix, that takes in either a row or column Matrix and then does something similar to the Map(...) method, where we can write a function that uses b_sub_i any way we want?