# A community convention for target function return values

**URL:** https://discourse.julialang.org/t/a-community-convention-for-target-function-return-values/42248
**Category:** General Usage
**Tags:** statistics, optimization, machine-learning
**Created:** [June 29, 2020, 1:03pm UTC](https://discourse.julialang.org/t/a-community-convention-for-target-function-return-values/42248 "2020-06-29T13:03:24Z")
**Posts on this page:** 1
**Showing post:** 13

<div class="post-metadata">

### Author: ![oschulz](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/oschulz/32/2998_2.png) [@oschulz](https://discourse.julialang.org/u/oschulz)
#### Post date: [July 1, 2020, 9:02am UTC](https://discourse.julialang.org/t/a-community-convention-for-target-function-return-values/42248/13 "2020-07-01T09:02:58Z")

</div>

> [@Tamas\_Papp](#):
>
> My point (which is well known) was that it is very easy to get (otherwise proper, well-behaved) posteriors where the “mode” is at a singularity.

Certainly, there are such cases. But we very often have physics applications where the model isn’t multi-level, or at least not with many levels, and where the posterior is well behaved. In such cases, it’s quite common in our community to determine the mode as well as credible regions and other quantity.

There are nasty cases too, of course … but many times, the real word is actually kind to us. 🙂

---

_[View the full topic](https://discourse.julialang.org/t/a-community-convention-for-target-function-return-values/42248)._