Teaching with Rewards and Punishments: Reinforcement or Communication?

40 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Interactive Learning from Policy-Dependent Human Feedback (2017, background): “ the feedback strategies they can use. Indeed, prior research has shown that interpreting human feedback like a reward function often induces positive reward cycles that leads to unintended behaviors [10, 11]. The issues with interpreting feedback as reward have led to the insight that human feedbac”
Interactively shaping robot behaviour with unlabeled human instructions (2019, methodology): “However, several authors have pointed out the difference between the nature of evaluative feedback and MDP rewards, considering them as information about the policy [15].”
Reinforcement learning in robotic applications: a comprehensive survey (2021, methodology): “Thus, the human-centered algorithm has been developed (Ho et al. 2015; Li et al. 2019).”