preference-based reinforcement learning

Preference-based reinforcement learning leverages user feedback or preferences instead of explicit reward signals to guide the learning process, allowing agents to adapt their behavior based on subjective evaluations of outcomes.

8 papers