Preference-based Reinforcement Learning beyond Pairwise Comparisons: Benefits of Multiple Options

Open in new window