Chad Scherrer

Right. I think this is yet another case where multiple dispatch can help us. Let's look at a simple case: Say you have parameter space $\Theta$ and observation space $X$....

Thanks @nignatiadis ! I see how the result in the normal case is equivalent to a convolution, but more generally I think `bind` builds a compound distribution. So for us,...

Some background on `bind`... In functional programming, a _monad_ is a type constructor `m` that has two operations: ``` unit :: a -> m a bind :: m a ->...

Good find! That looks very useful

Some design choices in Distributions can cause problems, like type constraints on constructors (making it unusable with Symbolics) and inconsistent argument checking in the form of the `check_args` keyword argument...

In the original `x ~ foo`, we can transport `foo` to `StdNormal() ^ n`. So the transport itself is deterministic, but the noise is in _what_ it's transporting. Hmm, I...

Thanks @sethaxen , the details here are really helpful. I guess to this point the marginalization would be an explicit sum? In some cases the algebraic form of this can...

Oh right, you already gave some concrete examples :)

Sure :) This one ```julia @inline function rand(d::For{F,T,D,X}) where {F,T

Yes, that's right. Also, I just realized the last example has the interesting property that it acts as a suspended computation: ```julia julia> θ = (sqrt(p) for p in 1:10)...