danielward27/flowjax

Better support for bounded/constrained distributions

Closed this issue · 0 comments

I am happy leaving it to the user to enforce constraints they require, rather than having explicitly constrained distributions, however:

  • We need a Log/Exp bijection
  • We should ensure that Distribution.log_prob calculations involving arctanh and log return -jnp.inf where appropriate (for samples out of support), rather than jnp.nan.
  • We should have an example of learning a distribution with bounded support