Verifiable reward
A verifiable reward is a reward signal whose correctness can be determined by an objective criterion or procedure.
Reinforcement Learning from Human Feedback
A verifiable reward is a reward signal whose correctness can be checked objectively by a procedure or criterion.
- How it is used
- Trains reasoning-oriented models on tasks whose completions can receive definitive correctness scores.
- How deep it goes
- Defines verifier-scored rewards and illustrates extraction plus binary scoring, while deferring detailed optimizer implementation.
- Examples
- Uses a prime-number summation checked by extracted answer, alongside coding, mathematics, and theorem-proving feedback signals.
- What you can do after
- Apply it via a verifier that extracts a completion outcome and assigns correctness-based reward.
Read how Reinforcement Learning from Human Feedback explains verifiable reward →
Build a Reasoning Model (From Scratch)
A reward signal that can be automatically checked against a correct answer or objective criterion.
- How it is used
- Evaluates math answers and supplies deterministic rewards for reinforcement-learning updates.
- How deep it goes
- Defines binary verification and the training loop, while deferring secure code-execution environments.
- Examples
- Uses calculator-like math checking, reference answers, binary correctness labels, and required boxed output.
- What you can do after
- Build a deterministic verifier that converts correct, extractable answers into training rewards.
Read how Build a Reasoning Model (From Scratch) explains verifiable reward →
Related concepts
- reinforcement learning with verifiable rewardsused by · Reinforcement learning with verifiable rewards is a training approach that optimizes a model using rewards derived from objectively checkable outcomes.
- Group Relative Policy Optimizationused by · Group Relative Policy Optimization is a reinforcement-learning algorithm that updates a policy using relative rewards or advantages estimated from groups of sampled outputs or actions.
- reward hackingcontrasts with · Reward hacking is behavior that exploits weaknesses or mismatches in a reward or evaluation specification to obtain a high measured score without achieving the intended objective.
- reinforcement learning with human feedbackcontrasts with · Reinforcement learning with human feedback is a model-training approach that uses human evaluations or preferences to guide optimization of model behavior.
- reinforcement learningused by · Reinforcement learning is a machine-learning paradigm in which an agent learns actions or a policy through interaction with an environment and evaluative feedback, such as rewards.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.