# Global optima with centralized critic (basic understanding)

**URL:** <https://discuss.ray.io/t/global-optima-with-centralized-critic-basic-understanding/1478>\
**Category:** RLlib\
**Created:** [March 30, 2021, 11:27pm UTC](https://discuss.ray.io/t/global-optima-with-centralized-critic-basic-understanding/1478 "2021-03-30T23:27:21Z")\
**Posts on this page:** 1\
**Showing post:** 10

<div class="post-metadata">

**Author:** ![aronium](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/aronium/32/1068_2.png) [@aronium](https://discuss.ray.io/u/aronium)\
**Post date:** [April 7, 2021, 8:36am UTC](https://discuss.ray.io/t/global-optima-with-centralized-critic-basic-understanding/1478/10 "2021-04-07T08:36:14Z")

</div>

This discusssion seems to be very similar to the one I started earlier this week. Maybe this is interesting for you, @CodingBurmer: [Multi-Agent System for maximizing the overall reward of all agents?](https://discuss.ray.io/t/multi-agent-system-for-maximizing-the-overall-reward-of-all-agents/1571) ?

---

_[View the full topic](https://discuss.ray.io/t/global-optima-with-centralized-critic-basic-understanding/1478)._