ePOWERAI
CoursesReinforcement Learning

Machine learning

Reinforcement Learning

Policy gradient, Q-learning, and actor-critic methods built from the Bellman equation up. Train agents in MuJoCo and Atari.

18 lessons · runs in your browser · no account

Your progress00 / 18 · 0%
4 lessons here are on the Guided Foundations Path· Steps 11-14

Course contents

Guided Path · Beginner

  1. W01Reinforcement Learning: Agents, Rewards, and MDPs
  2. W02Bandits: Exploration, Exploitation, and Regret
  3. W03Returns, Value, and the Bellman Equation
  4. W04Bandits: UCB1 and Thompson Sampling
  1. W01Week 1: Reinforcement Learning Problem Formulation
  2. W02Week 2: Multi-Armed Bandits
  3. W03Week 3: Dynamic Programming for Finite MDPs
  4. W04Week 4: Monte Carlo and Temporal-Difference Learning
  5. W05Week 5: Function Approximation in Reinforcement Learning
  6. W06Week 6: Deep Q-Learning and Variants
  7. W07Week 7: Policy Gradient and Actor–Critic Methods
  8. W08Week 8: Modern Deep Reinforcement Learning Algorithms
  9. W09Week 9: Exploration, Partial Observability, and Multi-Agent Reinforcement Learning
  10. W10Week 10: Model-Based Reinforcement Learning and Planning
  11. W11Week 11: Offline Reinforcement Learning
  12. W12Week 12: Reinforcement Learning from Human Feedback
  13. W13Week 13: Direct Preference Optimization and GRPO
  14. W14Week 14: Agentic Systems and Course Capstone