A Gallery of Methods Beyond RL — Part I: Sampling Methods | Shengyu Feng
shengyu-feng.github.io · 2,119 words · saved by 2 readers
A tour of methods beyond reinforcement learning.
Reinforcement learning (RL) has become a dominant paradigm for optimizing modern machine learning systems, from AlphaGo to large language models (LLMs). Given an evaluable objective, the recipe is simple: treat the (negative) objective as a reward, sample solutions from a policy, and optimize the expected reward. This naturally raises a question: Is reinforcement learning the only way to optimize an objective? The answer is clearly no. Many methods approach the same problem from very different perspectives. In this series, I want to collect some of these ideas, including some of my own…
saved by
related reading
- State of RL for reasoning LLMs | A. Weersaweers.de
- [2602.11399] Can We Really Learn One Representation to Optimize All Rewards?arxiv.org
- Sampling: Two Basic Algorithmsgregorygundersen.com
- RLHF | John Lambertjohnwlambert.github.io
- [2506.22401] Exploration from a Primal-Dual Lens: Value-Incentivized Actor-Critic Methods for Sample-Efficient Online RLarxiv.org
- RL_Notes__final_.pdfjubayer-ibn-hamid.github.io
- PPO for LLMs: A Guide for Normal Peoplecameronrwolfe.substack.com
- [1710.10044] Distributional Reinforcement Learning with Quantile Regressionarxiv.org
- GRPO++: Tricks for Making RL Actually Workcameronrwolfe.substack.com
- [2602.19362] LLMs Can Learn to Reason Via Off-Policy RLarxiv.org
- A vision researcher’s guide to some RL stuff: PPO & GRPO - Yuge (Jimmy) Shiyugeten.github.io
- RLHF: Reinforcement Learning from Human Feedbackhuyenchip.com