#Q-learning

When Q-learning Hyperparameters Are Dictated by Personality: A Reinforcement Learning Design Retrospective of a Civilization Simulator

A developer shares a civilization simulator where each agent's Q-learning hyperparameters (alpha, gamma) derive from personality traits, covering reward-shaping flaws, bootstrap masking, and stability risks.