Installs as a private draft. Your edits and self-improvement runs do not change the published bundle.
Best-of-N sampling with a reward model; simplest "RL" that actually works
Best-of-N sampling with a reward model; simplest "RL" that actually works
Sometimes you don't need fancy algorithms
RL Training Infrastructure > RL Algorithm Implementation
--- name: "Implement Rejection Sampling" version: "1.0.0" description: "Best-of-N sampling with a reward model; simplest \"RL\" that actually works" tags: - rl-skills-universe - "rl-training-infrastructure" - "rl-algorithm-implementation" trigger: "implement-rejection-sampling" --- # Implement Rejection Sampling Best-of-N sampling with a reward model; simplest "RL" that actually works ## Why It Matters Sometimes you don't need fancy algorithms ## Domain **RL Training Infrastructure** > RL Algorithm Implementation