Course
Reinforcement Learning
Learning to act from rewards, from Q-learning to RLHF.
L3 · AdvancedEvolvingKnown~3 h
What you’ll learn
- Define agents, rewards, policies, and the RL problem
- Explain value functions, Q-learning, and policy-gradient methods
- Describe exploration-exploitation and how RLHF aligns language models
Prerequisites
machine-learning
Module 1. RL Core
The RL problem, value functions, and Q-learning.
Module 2. Modern RL
Policy gradients, PPO, RLHF, and exploration.