All cursus
AIadvanced
🎯Reinforcement Learning today
A short, modern tour of where RL is actually working in 2026: the post-RLHF landscape, DPO and its variants, RL with verifiable rewards (GRPO, DeepSeek-R1 style), and how reasoning models use RL on chains of thought.
0 of 5 lessons complete
Sign in to track progress and earn a certificate.
Lessons in order
- 1AIReinforcement Learning in 2026: Where It Ships and Where It StallsStart
- 2AIRLHF: From Christiano 2017 to InstructGPT to the Offline EraStart
- 3AIDirect Preference Optimization: DPO, IPO, KTO, and SimPOStart
- 4AIRLVR and GRPO: Training LLMs with Verifiable RewardsStart
- 5AIRL in Reasoning Models: How o1, DeepSeek-R1, and Friends ThinkStart