SOURCE-LINKED INTELLIGENCE

Specifying Reward Functions for RL Without Environment Sampling

arXiv · AI, language, vision and robotics · article · Sep 14, 2026 · UTC

Enabling human stakeholders to specify reward functions that lead to their desired outcomes is a key challenge in deploying reinforcement learning agents. Preference-based methods such as online RLHF can reduce the burden of manual reward design, but they require repeatedly training policies, sampling trajectories from the real world, and eliciting feedback, making them impractical in settings where environment interaction is computationally expensive or unsafe. We introduce Experience-Free Autonomous Reward Specification (EARS), a method for learning reward functions from preferences without

Read original source ↗ Open in workspace

recordType

paper

region

Global