search

marlbenchmark / on-policy

This is the official implementation of Multi-Agent PPO (MAPPO).
https://sites.google.com/view/mappo
MIT License
1.27k stars 292 forks source link

Does the state of the grid world need to be normalized? #105

Closed saiyuhang123 closed 2 months ago

saiyuhang123 commented 5 months ago

I have made a multi-agent pathfinding in a grid world. The state adopts the Manhattan distance between the agent and the goal. Does it need to be normalized when inputting the state information to mappo?

zoeyuchao commented 4 months ago

Generally you don‘t need to do normalization by yourself since we will do valuenorm when processing the observation in the policy network.