2 - Deep RL and RL post-training intro

Second lecture for CSE 599J on Social Reinforcement Learning: https://courses.cs.washington.edu/cou.... Extremely fast intro to deep reinforcement learning algorithms, covering popular off-policy and on-policy algorithms and how to debug them. Then gets into RL post-training of language models, and a brief intro to RL from Human Feedback (RLHF).