This research explores reinforcement learning with human feedback for constructing spanning structures. In this framework, individuals compare two demonstrations of a task and select their preferred option, as shown in the video. Subsequently, an agent learns to autonomously build bridges from these pairwise comparisons.
