Fritz Obermeyer

Kevin Murphy recommended a paper by the ForneyLab authors > In the case where discretes are children of Gaussians (eg Gaussian state space models which generate discrete observations), you have...

@eb8680 interesting... I guess we could even enforce constraints, something like ```py def assume(constraint, value): if not constraint: raise TypeError return value @function def matmul( x: Dependent[lambda x: assume(isinstance(x, Array)...

@ordabayevy yes I think your `Log` functor relates the semiring pair `((add,mul), (logaddexp,add))` and the pair `((max,mul),(max,add))`. However no functor relates e.g. the pair `((add,mul), (max,add))`. Indeed `Funsor`s can represent...

cc @martinjankowiak

@fehiepsi Do you have time to look into a pyro.ops.studentt analogous to pyro.ops.gaussian? I think a `pyro.distributions.StudentTHMM` would be a very competitive time series model if we can make the...

I think we could do this with an auxiliary variable reparameterizer like `pyro.poutine.reparam`. In funsor I believe this would be a `reparam` interpretation that could e.g. transform a `StudentT` or...

@eb8680 great questions! 1. I like the idea of being not only more Pythonic but closer to the NumPy ecosystem. Let's see what we can find. EDIT it looks like...

@eb8680 suggested this will be easier after #491 , e.g. we could add an `@ops.make` decorator and use it in the VAE example e.g. ```py ... encoder = Encoder() decoder...

@jpchen I could explain details you if you're interested

Re: coercion, here's another idea. We could allow `Bint[n]` to include any number in `{-n, ..., n-1}` where under indexing any `k>0` is equivalent to `n-k`. Then coercing from `n`...