Machine Learning
Reinforcement Learning
Learning by trial and reward. Agents, environments, Q-learning and policy gradients, with runnable grid-world examples.
- Advanced
- About 14 hours
- 3 modules
- 12 lessons
- Free
This course is not released yet. We publish courses one at a time, each one complete. The full syllabus is below.
What you will learn
- FoundationsAgents, Environments and Rewards, Markov Decision Processes, Exploration vs Exploitation and more
- Classic MethodsMulti-Armed Bandits, Q-Learning, SARSA and more
- Deep RLDeep Q-Networks, Policy Gradients, Actor-Critic Methods and more
Course syllabus
Not released yet · full syllabus below
- 1
Foundations
4 lessons- 1Agents, Environments and RewardsSoon
- 2Markov Decision ProcessesSoon
- 3Exploration vs ExploitationSoon
- 4Value FunctionsSoon
- 2
Classic Methods
4 lessons- 5Multi-Armed BanditsSoon
- 6Q-LearningSoon
- 7SARSASoon
- 8Solving a Grid WorldSoon
- 3
Deep RL
4 lessons- 9Deep Q-NetworksSoon
- 10Policy GradientsSoon
- 11Actor-Critic MethodsSoon
- 12Where RL Is UsedSoon