ForagerRL - gama-platform/gama GitHub Wiki
By Killian Trouillet
Welcome to the comprehensive tutorial on Reinforcement Learning with the GAMA platform. You will build a forager agent that learns to navigate toward food while avoiding obstacles — from a simple grid world to a continuous environment trained with Deep RL.
Build a tabular Q-Learning agent entirely in GAML, step by step:
- Step 1: The Grid World — Create the 10×10 environment with food and obstacles.
- Step 2: The Forager Agent — Define a simple agent that moves randomly.
- Step 3: Rewards and Episodes — Implement the reward system and simulation resets.
-
Step 4: The Q-Table — Set up the agent's memory using
map<string, float>. - Step 5: Q-Learning Algorithm — Implement the Bellman equation and ε-greedy policy.
- Step 6: Visualization & Automatic Test — Add charts, heatmaps, and evaluate the learned policy.
