Sam Daulton

Hi @davifebba, sorry for the delay. That looks good, but "end_time" should be the current time, not the elapsed time since the start.

Yeah, that would be nice. Using box decompositions (`DominatedPartitioning`) to compute hypervolumes seems to work more efficiently than the dimension-sweep algorithm, empirically. Both would likely benefit from C++

The current implementation is not differentiable and is simply used for evaluation purposes.

I added [this rule to `is_non_dominated`](https://github.com/pytorch/botorch/blob/5012fe8a39b434e1b0f3d3a968eb17b3dd0c9e27/botorch/utils/multi_objective/pareto.py#L47-L48) to avoid memory issues and very long run-times on cpu. It looks like that slows things down quite a bit on a GPU. I...

> Note that the dataset that I used comes with the moocore package as well (it is documented [here](https://multi-objective.github.io/moocore/r/reference/CPFs.html), I am not sure how to document datasets in Python), so...

> How does it compare with [moocore.is_nondominated](https://multi-objective.github.io/moocore/python/reference/generated/moocore.is_nondominated.html#moocore.is_nondominated) ? I didn't explicitly test moocore, since I didn't have it installed, but I would imagine they are at least competitive on a...

This makes sense to me. An alternative to passing `pyro_model_kwargs` would be to specify the specific args that you'd like to pass, to allow for stricter input checking. What kwargs...

Sorry for the delay here. I lost track of this issue. A couple thoughts: 1. If we just pass `pyro_model_kwargs`, * Then we have little type-checking/validation of the arguments, which...

Thanks for pointing this out. Would you mind putting up a PR to make it consistent?

@yucenli had worked on implementing this. I can clean that up and put up a PR