Ecological Public Good

Class for ecological public good environment

The environment was introduced in

Barfuss, W., Donges, J. F., Vasconcelos, V. V., Kurths, J., & Levin, S. A. (2020). Caring for the future can turn tragedy into comedy for long-term collective action under risk of collapse. Proceedings of the National Academy of Sciences, 117(23), 12915-12922.

Example

from pyCRLD.Environments.EcologicalPublicGood import EcologicalPublicGood
from pyCRLD.Agents.StrategyActorCritic import stratAC
from pyCRLD.Utils import FlowPlot as fp
import numpy as np

env = EcologicalPublicGood(N=2, f=1.2, c=5, m=-4, qc=0.2, qr=0.1, degraded_choice=True)
env

EcologicalPublicGood_2_1.2_5_-4_0.2_0.1_DegChoi

In the prosperous state, the rewards are a tragedy Prisoners’ Dilemma.

env.R[0,1,:,:,1], env.R[1,1,:,:,1]

(array([[ 1., -2.],
        [ 3.,  0.]]),
 array([[ 1.,  3.],
        [-2.,  0.]]))

Yet, because of the possible collapse and the agents’ future outlook, the overall regime is one of coordination.

# Init enviornment and MultiAgentEnvironment-interface
mae = stratAC(env=env, learning_rates=0.1, discount_factors=0.9)

x = ([0], [0,1], [0])  # Plotting on the x-axis the [0]'s agents probability in both states [0,1] to cooprate [0]
y = ([1], [0,1], [0])  # Plotting on the y-axis the [1]'s agents probability in both states [0,1] to cooprate [0]
ax = fp.plot_strategy_flow(mae, x, y, flowarrow_points = np.linspace(0.01 ,0.99, 9), NrRandom=16)

Implementation

---

source

EcologicalPublicGood

 EcologicalPublicGood (N:int, f:Union[float,Iterable],
                       c:Union[float,Iterable], m:Union[float,Iterable],
                       qc:Union[float,Iterable], qr:Union[float,Iterable],
                       degraded_choice=False)

Ecological Public Good Environment.

	Type	Default	Details
N	int		number of agents
f	Union		public goods synergy factor
c	Union		cost of cooperation
m	Union		collapse impact
qc	Union		collapse leverage/timescale
qr	Union		recovery leverage/timescale
degraded_choice	bool	False	whether agents have a choice at the degraded state

---

source

EcologicalPublicGood.actions

 EcologicalPublicGood.actions ()

The action sets

---

source

EcologicalPublicGood.states

 EcologicalPublicGood.states ()

The states set

---

source

EcologicalPublicGood.TransitionTensor

 EcologicalPublicGood.TransitionTensor ()

Get the Transition Tensor.

The TransitionTensor is obtained with the help of the _transition_probability method.

---

source

EcologicalPublicGood._transition_probability

 EcologicalPublicGood._transition_probability (s:int, jA:Iterable, s_:int)

Returns the transition probability for current state s, joint action jA, and next state s_.

	Type	Details
s	int	the state index
jA	Iterable	indices for joint actions
s_	int	the next-state index
Returns	float	transition probability

---

source

EcologicalPublicGood.RewardTensor

 EcologicalPublicGood.RewardTensor ()

Get the Reward Tensor R[i,s,a1,…,aN,s’].

The RewardTensor is obtained with the help of the _reward method.

---

source

EcologicalPublicGood._reward

 EcologicalPublicGood._reward (i:int, s:int, jA:Iterable, s_:int)

Returns the reward value for agent i in current state s, under joint action jA, when transitioning to next state s_.

	Type	Details
i	int	the agent index
s	int	the state index
jA	Iterable	indices for joint actions
s_	int	the next-state index
Returns	float	reward value

---

source

EcologicalPublicGood.id

 EcologicalPublicGood.id ()

Returns id string of environment