# Multi agent unique actions
**URL:** https://discuss.ray.io/t/multi-agent-unique-actions/14717
**Category:** Configure Algorithm, Training, Evaluation, Scaling
**Created:** [May 16, 2024, 9:58pm UTC](https://discuss.ray.io/t/multi-agent-unique-actions/14717 "2024-05-16T21:58:21Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@jjgriffin2](https://discuss.ray.io/u/jjgriffin2)
#### Post date: [May 16, 2024, 9:58pm UTC](https://discuss.ray.io/t/multi-agent-unique-actions/14717/1 "2024-05-16T21:58:21Z")
Finding a solution to the following is critical for project completion.
I have a project that is multiagent where each agent is identical an and has its own mapped policy. The action space is discrete, and the environment is a custom, shared multi agent environment.
The step routine receives, as expected, an action\_dict that maps agents to actions.
Is there a way to intercept the creation of the action\_dict during manual training (via algo.train, where the algorithm is PPO) such that the actions among agents are unique? I can’t seem to locate where in the code the action dictionary is created, and am clueless as to how to override the standard implementation. AA this
TIA
Jim Griffin
---
### Author:  [@zoe\_tsekas](https://discuss.ray.io/u/zoe_tsekas)
#### Post date: [May 29, 2024, 1:37pm UTC](https://discuss.ray.io/t/multi-agent-unique-actions/14717/2 "2024-05-29T13:37:41Z")
>