What you'll learn

Many aspiring machine learning engineers struggle to move from understanding reinforcement learning theory to applying it in code, especially when scaling to complex problems. In this course, Introduction to Q-learning, you’ll learn to implement both traditional and deep Q-learning to train intelligent agents. First, you’ll explore the foundations of Q-learning, its differences from other RL methods like SARSA, and the role of Q-functions and Q-tables. Next, you’ll discover how to build a deep Q-network that approximates Q-values using neural networks and updates using gradient descent. Finally, you’ll learn how to train your Q-network in Gym environments, use experience replay and target networks, and monitor learning over time. When you’re finished with this course, you’ll have the skills and knowledge of Q-learning needed to build scalable agents that learn from experience.