Tag
OpenAI demonstrates fine-tuning GPT-2 (774M parameters) using human preference feedback for text continuation and summarization tasks, requiring 5k labels for stylistic tasks and 60k for summarization, with models achieving 86-88% human preference rates though revealing labeler heuristic exploitation.
OpenAI presents iterated amplification, a method for training AI systems on complex tasks by recursively decomposing them into smaller subtasks that humans can judge and solve, building up training signals from scratch through iterative composition.
OpenAI releases RL-Teacher, an open-source tool for training AI systems through human feedback instead of hand-crafted reward functions, with applications to safe AI development and complex reinforcement learning problems.
OpenAI presents a method for training AI agents using human preference feedback, where an agent learns reward functions from human comparisons of behavior trajectories and uses reinforcement learning to optimize for the inferred goals. The approach demonstrates strong sample efficiency, requiring less than 1000 bits of human feedback to train an agent to perform a backflip.