A social reinforcement learning agent

By Charles Lee Isbell Jr., Christian R. Shelton, Michael Kearns, Satinder Singh, Peter Stone. 2001

170 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Sociology and Political Science

How researchers see this paper

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

In Artificial Intelligence

In Sociology and Political Science

Citations in the corpus

Leveraging Human Guidance for Deep Reinforcement Learning Tasks (2019, background): “In contrast to policy shaping, reward shaping interprets human feedback as the value function or a human-speciﬁed reward function [Isbell et al. , 2001; Tenorio-Gonzalez et al. , 2010; Pilarski et al. , 2011].”
Interactive Learning from Policy-Dependent Human Feedback (2017, background): “For example, it was often observed that trainers taper their feedback over the course of learning (Ho et al., 2015; Knox et al., 2012; Isbell et al., 2001).”
Reinforcement Learning for Bandit Neural Machine Translation with Simulated Human Feedback (2017, neutral): “, 2011), and virtual assistants (Isbell et al., 2001).”
Learning something from nothing: Leveraging implicit human feedback strategies (2014, methodology): “We used two learning algorithms (SABL and ISABL) [11] that inferred the correct behavior using our feedback model, as well as two algorithms (M−0, M+0) based on algorithms in the literature on maximizing human feedback considered as reward [6], [8].”
Balancing Multiple Sources of Reward in Reinforcement Learning (2000, background): “If the users of the system know they are training it, they will employ all kinds of reward strategies to try to steer the system to the desired behavior [2].”
Learning behaviors via human-delivered discrete feedback: modeling implicit feedback strategies to speed up learning (2015, unknown): “M+0 is designed to be similar to the COBOT system [11] in how it handles episodes without feedback, treating no feedback as a reward value of zero.”
Reinforcement Learning with Human Teachers: Evidence of Feedback and Guidance with Implications for Learning Performance (2006, methodology): “…approaches (e.g., the human contributes to the reward function, or supervises the action selection) are positioned strongly along the exploration dimension (Blumberg et al. 2002; Kaplan et al. 2002; Isbell et al. 2001; Kuhlmann et al. 2004; Evans 2002; Clouse & Utgoff 1992; Smart & Kaelbling 2002).”
DQN-TAMER: Human-in-the-Loop Reinforcement Learning with Intractable Feedback (2018, background): “It is also reported that the feedback frequency varies largely among human users [19], [20].”
Interactively shaping robot behaviour with unlabeled human instructions (2019, methodology): “In the standard method, reward shaping, evaluative feedback is converted into numerical values that are used for augmenting a predefined reward function [16, 30, 43, 47].”
FRESH: Interactive Reward Shaping in High-Dimensional State Spaces using Human Feedback (2020, background): “Reward shaping modifies rewards supplied by the environment to accelerate the learning process of the agent [5, 17, 41].”
The EMPATHIC Framework for Task Learning from Implicit Human Feedback (2020, background): “[21] C.”
PERCEPTUAL GOAL SPECIFICATIONS FOR REINFORCEMENT LEARNING (2017, background): “For example, some approaches allow humans to provide rewards to agents in real-time [27, 35, 60].”
Policy Shaping: Integrating Human Feedback with Reinforcement Learning (2013, background): “F or example, adding a drift parameter to account for the human tendency to give less feedback over t im [1, 12].”
Teachable robots: Understanding human teaching behavior to build more effective robot learners (2008, background): “Informed by related work (Isbell et al., 2001), it is reasonable to expect people would habituate to the activity and that feedback would decrease over the training session.”
The rise and potential of large language model based agents: a survey (2025, unknown): “…capabilities and data availability, along with a growing interest in simulating interactions between intelligent agents and their environments, researchers have begun to utilize reinforcement learning methods to train agents for tackling more challenging and complex tasks [11,12,44,45].”