Reinforcement Learning From Human Feedback

concepts · 4 notes linked

Related: Anthropic · AI Safety · Openai · Chatgpt · AI Alignment · Deepmind · Paul Christiano · Large Language Models

Notes

• 255 — Paul Christiano's defense of RLHF research positive net impact
• Perspectives on the Social Impacts of Reinforcement Learning with Human Feedback — Social and ethical impacts of RLHF across seven societal dimensions
• The Capacity for Moral Self-Correction in Large Language Models — RLHF-trained LLMs can self-correct harmful outputs when instructed
• The challenges of reinforcement learning from human feedback (RLHF) - TechTalks — RLHF limitations across feedback, reward modeling, and policy