Predator Prey
These policies are for the Predator Prey environment. Read environment page for detailed information about the environment.
Generic
These policies can be used for any version of this environment.
env = posggym.make("PredatorPrey-v0")
Policy |
ID |
Valid Agent IDs |
Description |
|---|---|---|---|
|
|
All |
H1 moves towards closest observed prey, closest observed predator, or explores randomly, in that order. |
|
|
All |
H2 moves towards closest observed prey, closest observed predator, or explores in a clockwise spiral around arena, in that order. |
|
|
All |
H3 moves towards closest observed prey to the closest observed predator or explores in a clockwise spiral around arena, in that order. |
grid=10x10-num_predators=2-num_prey=3-cooperative=True
env = posggym.make(
"PredatorPrey-v0",
grid="10x10",
num_predators=2,
num_prey=3,
cooperative=True,
prey_strength=2,
obs_dim=2
)
Policy |
ID |
Valid Agent IDs |
Description |
|---|---|---|---|
|
|
All |
Deep RL policy trained using PPO and self-play. |
|
|
All |
Deep RL policy trained using PPO and self-play. |
|
|
All |
Deep RL policy trained using PPO and self-play. |
|
|
All |
Deep RL policy trained using PPO and self-play. |
|
|
All |
Deep RL policy trained using PPO and self-play. |
|
|
All |
Deep RL policy trained using PPO and self-play. |
|
|
All |
Deep RL policy trained using PPO and self-play. |
|
|
All |
Deep RL policy trained using PPO and self-play. |