Reviews: Regret Minimization for Reinforcement Learning with Vectorial Feedback and Complex Objectives