Online human training of a myoelectric prosthesis controller via actor-critic reinforcement learning

By Patrick M. Pilarski, Michael Rory Dawson, Thomas Degris, F. Fahimi, Jason P. Carey, Richard S. Sutton. 2011

185 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Cellular and Molecular Neuroscience, Cognitive Neuroscience, Reinforcement Learning

How researchers see this paper

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

In Biomedical Engineering

1. Sergey Levine
2. Pieter Abbeel
3. Laura M. Smith
4. Scott Niekum
5. Kimin Lee
6. Peter Stone
7. Anca Dragan
8. W. B. Knox

In Cellular and Molecular Neuroscience

1. Borja Ibarz

In Cognitive Neuroscience

1. Sergey Levine
2. Pieter Abbeel
3. Anca Dragan
4. Scott Niekum
5. Peter Stone
6. Kimin Lee
7. W. B. Knox
8. Laura M. Smith

In Reinforcement Learning

1. Sergey Levine
2. Pieter Abbeel
3. Scott Niekum
4. Anca Dragan
5. Peter Stone
6. Chelsea Finn
7. Jan Leike
8. Shane Legg

Citations in the corpus

• Deep Reinforcement Learning from Human Preferences (2017, background): “A long line of work studies reinforcement learning from human ratings or rankings, including Akrour et al. (2011), Pilarski et al. (2011), Akrour et al. (2012), Wilson et al. (2012), Sugiyama et al. (2012), Wirth and Fürnkranz (2013), Daniel et al. (2015), El Asri et al. (2016), Wang et al. (2016),…”
• Reward learning from human preferences and demonstrations in Atari (2018, background): “There is a large body of work on reinforcement learning from human ratings or rankings (Wirth et al., 2017): Knox and Stone (2009), Pilarski et al. (2011), Akrour et al. (2012), Wilson et al. (2012), Wirth and Fürnkranz (2013), Daniel et al. (2015), El Asri et al. (2016), Wirth et al. (2016),…”
• Supervising strong learners by amplifying weak experts (2018, background): “…box (RL from a human training signal) is beginning to be explored, and includes inverse reinforcement learning (Ng and Russell, 2000; Abbeel and Ng, 2004; Finn et al., 2016) and RL from human feedback (Knox and Stone, 2009; Pilarski et al., 2011; MacGlashan et al., 2017; Christiano et al., 2017).”
• Reliability and Learnability of Human Bandit Feedback for Sequence-to-Sequence Reinforcement Learning (2018, methodology): “The actor-critic framework has been adapted to interactive RL from human feedback by Pilarski et al. (2011) and MacGlashan et al. (2017).”
• Shared Autonomy via Deep Reinforcement Learning (2018, background): “ontrol and reinforcement learning algorithms to implement closed-loop decoder adaptation [29] for applications like prosthetic limb controllers that respond to neural signals from myoelectric sensors [25]. These algorithms typically track desired motion, whereas we focus on tasks with long-horizon g”
• Policy Shaping: Integrating Human Feedback with Reinforcement Learning (2013, methodology): “This is the most common way human feedback has been applied to RL [1–5].”
• Learning non-myopically from human-generated reward (2013, background): “This mapping to 1 and -1, though not infallible, is an intuitive choice that is similar to that of related works that explain their exact mappings [14, 13, 10, 11].”
• Model-Free reinforcement learning with continuous action in practice (2012, unknown): “[18], we do not think it can scale to high-dimension action space as such.”
• Training an Actor-Critic Reinforcement Learning Controller for Arm Movement Using Human-Generated Rewards (2017, unknown): “[41] to successfully complete a 2-joint velocity”
• The EMPATHIC Framework for Task Learning from Implicit Human Feedback (2020, background): “[22] P.”