How Humans Teach Agents - A New Experimental Perspective

By W. B. Knox, Brian D. Glass, Bradley C. Love, W. Maddox, Peter Stone. 2012

85 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.

Expressing Arbitrary Reward Functions as Potential-Based Advice (2015, background): “On the other hand, there has been a lot of research in human-provided advice (Thomaz and Breazeal 2006), (Knox et al. 2012).”
Actor-Critic Reinforcement Learning with Simultaneous Human Control and Feedback (2017, background): “It has been shown that the limited human feedback can be applied across near-optimal state-action pairs, and support the agent learning an optimal solution (Pilarski et al., 2011; Knox et al., 2012).”
Learning behaviors via human-delivered discrete feedback: modeling implicit feedback strategies to speed up learning (2015, unknown): “Similar to thiswork, other studies [16] have examined howuserswant to provide feedback, finding that: (1) there is little difference in a trainer’s feedback whether they think that the agent can learn or that they are critiquing a fixed performance; and (2) humans can reduce the amount of feedback t”
Interactive reinforcement learning for task-oriented dialogue management (2016, background): “The user is usually willing to provide explicit or implicit feedback about the system’s actions if it leads to demonstrable improvements in the system’s performance (Knox et al. (2012)).”
Learning something from nothing: Leveraging implicit human feedback strategies (2014, background): “More similar to our work, Knox et al. [9] looks at how people want to provide feedback to learning agents.”