Kyurae Kim
Kyurae Kim
Betanalpha does [discuss](https://discourse.mc-stan.org/t/jacobian-of-softmax-tansformation-of-a-n-1-degrees-of-freedom-unbounded-parameter/18780/26?u=red-portal) a bijective softmax by arbitrarily setting the endpoint logits. Any experience with this?
Good to know thanks. Though, back to my original intention, I really wish that our simplex bijector could play nicely with GPUs out of the box. Among non-NF bijectors, it...
Actually, nevermind. I just wrote a stick-breaking bijector using array operations based on the implementations of numpyro and tensorflow. If this were to be added to `Bijectors.jl` we'll probably have...
I do have the feeling that this will have to wait until the batch operation interface is finalized. @torfjelde Do we have an expectation on when that would be?
Let me ping @yebai for good measure. Hong, are you happy if I go forward with this?
@sunxd3 @mhauru @yebai Could we move this forward?
> Small technical question: am I reading it correctly that AdvancedVI right now uses the linear parametrization? Yes, the default settings do, hence the involvement of `ClipScale` or `ProximalLocationScaleEntropy`, but...
Hmmm... seems like `mapreduce` with Zygote is broken again.
Not sure about this one. Maybe @torfjelde @willtebbutt have more insight?
I'll close this issue for now. Feel free to re-open this if you find that more discussion is needed.