
Introduction to Reinforcement Learning (Lecture 1)
Emma Brunskill, Associate Professor of Computer Science at Stanford, opens CS234 with a full introduction to reinforcement learning. She lays out what distinguishes RL from supervised learning and other machine learning paradigms, emphasizing sequential decision making under uncertainty, delayed reward, and learning through interaction rather than labeled examples. The lecture covers the core components of an RL problem, including agents, environments, states, actions, and rewards, and previews the mathematical framework of Markov decision processes used throughout the course. Brunskill discusses real-world applications, from healthcare to robotics to game playing, and situates RL historically alongside milestones like AlphaGo. At 79 minutes, this is a standard first lecture: motivating the field, setting course expectations, and establishing the vocabulary and notation that later lectures build on.