Reinforcement Learning I (Intro Lecture)
Reinforcement Learning I (Intro Lecture)
This lecture provides an introduction to a variety of topics in Reinforcement Learning.
Topics covered in this lesson
- An overview of the features of a Reinforcement Learning system and general methods for predicting state values, including Monte Carlo methods, Dynamic Programming, and Temporal Differences learning
- An overview of the control problem in Reinforcement Learning, including the Bellman optimality equation, algorithms such as value iteration, Q-learning, SARSA, and methods for exploration
- A brief introduction to function approximation and deep RL
- Policy-based methods such as policy iteration, the actor-critic architecture, and the pros and cons of value-based vs. policy-based methods
Prerequisites
Experience with Python Programming Language.
Back to the course