Deep Reinforcement Learning from Human Preferences
by OpenAI · free · Last verified 2026-03-17
This foundational RLHF paper shows that human preference comparisons between agent behaviors can train a reward model that guides deep RL agents in complex tasks like Atari games and MuJoCo locomotion, without hand-crafted reward functions. The approach reduces human labeling effort by ~3 orders of magnitude compared to direct reward specification.
https://arxiv.org/abs/1706.03741 ↗B+
B+—Good
Adoption: AQuality: A+Freshness: C+Citations: A+Engagement: F
Specifications
- License
- Open Access
- Pricing
- free
- Capabilities
- reward-learning, preference-learning, rlhf, human-in-the-loop
- Integrations
- Use Cases
- llm-alignment, agent-training, reward-model-training
- API Available
- No
- Tags
- reinforcement-learning, rlhf, human-feedback, reward-learning, alignment
- Added
- 2026-03-17
- Completeness
- 100%
Index Score
78Adoption
88
Quality
95
Freshness
58
Citations
95
Engagement
0
Put AI to work for your business
Deploy this paper alongside autonomous AaaS agents that handle tasks end-to-end — no babysitting required.