Reinforcement learning improves behaviour from evaluative feedback

By Michael L. Littman. 2015

348 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Management Science and Operations Research

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 Management Science and Operations Research

Citations in the corpus

Reinforcement Learning in Healthcare: A Survey (2019, background): “games, robotics control, financial and business management, autonomous driving, natural language processing, computer vision, biological data analysis, and art creation, just to name a few [10], [11], [12], [13], [14].”
A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning (2017, background): “Deep reinforcement learning combines deep learning [59] with reinforcement learning [94, 64] to compute a policy used to drive decision-making [73, 72].”
A Tutorial on Thompson Sampling (2017, background): “Reinforcement learning (RL) extends upon contextual online decision problems to allow for delayed feedback and long term consequences [31, 32].”
Interactive machine learning for health informatics: when do we need the human-in-the-loop? (2016, background): “Driven by the increasing availability of rich data, RL has achieved great results, including developments in fundamental ML-relevant areas, such as generalization, planning, exploration, and empirical methodology, leading to better applicability to real-world problems [46].”
A Q-values Sharing Framework for Multiple Independent Q-learners (2019, neutral): “Reinforcement Learning (RL) [6] is successfully employed in many practical applications, such as robotics [5].”
Society-in-the-loop: programming the algorithmic social contract (2017, background): “Prominent examples include machines learning their own representations of the world via Deep Neural Network architectures [41], Reinforcement Learning from evaluative feedback [45], and economic reasoning in markets and other multi-agent systems [58].”
Adaptive Dynamic Programming for Control: A Survey and Recent Advances (2021, background): “Compared to ADP/ACD, the area of RL is more mature and has a vast amount of literature (see [16], [21]–[23]).”
A comprehensive survey on model compression and acceleration (2020, methodology): “To learn the best policy for pruning, the authors trained an agent based on reinforcement learning (Littman 2015), and each convolutional layer pruning is modelled as a Markov decision process (Puterman 2014).”
Deep Reinforcement Learning: An Overview (2017, neutral): “…reinforcement learning (Sutton and Barto, 2017; Szepesv´ari, 2010; Bertsekas, 2012; Powell, 2011; Bertsekas and Tsitsiklis, 1996; Puterman, 2005; Littman, 2015; Kaelbling et al., 1996); deep learning (LeCun et al., 2015; Goodfellow et al., 2016; Bengio, 2009; Deng and Dong, 2014); machine…”
Reinforcement learning in robotic applications: a comprehensive survey (2021, neutral): “2017; Littman 2015; Yang and Gu 2004; Luo et al.”
Control of synaptic plasticity in deep cortical networks (2018, positive): “Δwi,j = β ∙ fi(ai) ∙ fj(aj) ∙ RPE (2).”
Part-Activated Deep Reinforcement Learning for Action Prediction (2018, methodology): “Deep reinforcement learning technique is introduced to optimize the sequential model with delayed reward [23] and performs very promising results in a series of problems.”
Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition (2018, background): “For clarity, We use the bold-sized word to represent the actions of MDP.”
Attention-Aware Deep Reinforcement Learning for Video Face Recognition (2017, background): “Deep Reinforcement Learning: Reinforcement learning has been originated from our understanding of humans’ decision making process [25], which aims to enable the agent to decide the behavior from its experiences.”