02-rl-reward

RL Reward Construction with OpenJudge

Build reward signals for reinforcement learning from human feedback (RLHF) and reinforcement learning from AI feedback (RLAIF) using the openjudge library.

When to Use This Skill

• Building scalar rewards for GRPO / REINFORCE rollout scoring
• Generating (chosen, rejected) preference pairs for DPO / IPO
• Best-of-N candidate selection
• Multi-dimensional reward shaping (correctness + safety + format)
• Replacing or bootstrapping a reward model with LLM-as-judge

Step 1 — Choose Your Reward Strategy

Use this decision tree before writing any code:

RL Algorithm + Task type?
│
├── GRPO / REINFORCE — Verifiable task (math, code, structured output)
│   └── → POINTWISE  ✅  (FunctionGrader, exact score, zero LLM cost)
│
├── GRPO / REINFORCE — Subjective task (instruction following, dialogue, summarization)
│   └── → PAIRWISE TOURNAMENT  ✅  (compare each rollout vs all others in group,
│                                    reward = net win rate within group)
│
├── DPO / IPO / SLiC — need (chosen, rejected) pairs
│   └── → PAIRWISE  ✅  (two-way comparison, return winner/loser)
│
└── Best-of-N / reranking — rank N candidates
    └── → LISTWISE  ✅  (single call ranks all N at once)

Cost constraint?
├── Low budget
│   └── FunctionGrader (free) → pointwise; or pairwise with small judge model
│
├── Medium budget
│   └── Pointwise: 2–3 LLM graders + WeightedSumAggregator
│   └── Pairwise tournament: 1 LLM judge, N*(N-1)/2 comparisons per group
│
└── High quality / no cost limit
    └── Pointwise voting (3–5 calls) or pairwise with strong judge + debiasing

Sub-documents — Read When Relevant

Topic	File	Read when…
Pointwise multi-dim reward	pointwise.md	GRPO on verifiable tasks; multi-dimension scoring
Pairwise reward	pairwise.md	GRPO on subjective tasks (tournament); DPO/RLAIF preference pairs

Read the relevant sub-document before writing any code.