Aligning Human Preferences with Baseline Objectives in Reinforcement Learning.

Daniel Marta, Simon Holk, Christian Pek, Jana Tumova, Iolanda Leite

Browse the full ICRA paper archive.