A rigorous, foundational walkthrough of core RL algorithms (SARSA, Q-learning, policy gradients) that grounds the concepts increasingly relevant to agent training and RLHF-style fine-tuning, with clear explanations of how the methods actually differ.
[Updated on 2020-09-03: Updated the algorithm of SARSA and Q-learning so that the difference is more pronounced. [Updated on 2021-09-19: Thanks to 爱吃猫的鱼, we have this post in Chinese ].
Transcript
[Updated on 2020-09-03: Updated the algorithm of SARSA and Q-learning so that the difference is more pronounced. [Updated on 2021-09-19: Thanks to 爱吃猫的鱼, we have this post in Chinese ].
