20251230-KARL: Agents play the knowledge agreement game and adapt their knowledge base by selecting adaptation operators through Reinforcement Learning

This archive contains the results of a computational cultural knowledge evolution experiment [1,2].

Experiment design

Agents are playing the knowledge-based agreement game using their symbolic knowledge base.
Agents adapt their knowledge base by selecting operators through Reinforcement Learning.

Hypotheses: [Learned policies are independent of the environment, Learned policies enable agents to reach consensus efficiently]

Experimental Plan

The independent variables have been varied as follows:  
LEARNING_METHOD = [random+, thompson, softmax, a3c]  
SEED = [24, 25, 27, 2626, 2727] (randomly selected)

[1] https://sake.re/20251230-KARL
[2] https://gitlab.inria.fr/moex/karlOperators