Multi-Agent Reinforcement Learning: Foundations and Modern Approaches