XAI: Learning Fairness with Interpretable Machine