Robot learning from demonstration by constructing skill trees

By George Dimitri Konidaris, Scott Kuindersma, R. Grupen, Andrew Barto. 2012

358 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering, Representation & Self-Supervised Learning, Robotics & Control

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 Control and Systems Engineering

In Representation & Self-Supervised Learning

In Robotics & Control

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, methodology): “Konidaris et al. (2011a, 2012) proposes an approach that constructs a skill tree from human demonstrations.”
Learning human activities and object affordances from RGB-D videos (2012, background): “There is some recent work in interpreting human actions and interaction with objects (Lopes and Santos-Victor, 2005; Saxena et al., 2008; Aksoy et al., 2011; Konidaris et al., 2012; Li et al., 2012) in context of learning to perform actions from demonstrations.”
End-to-End Driving Via Conditional Imitation Learning (2017, background): “Basic motor skills that are learned in this framework can be transferred across tasks [19].”
Modular Multitask Reinforcement Learning with Policy Sketches (2016, methodology): “An alternative family of approaches employs post-hoc analysis of demonstrations or pretrained policies to extract reusable sub-components (Stolle & Pre-cup, 2002; Konidaris et al., 2011; Niekum et al., 2015).”
An Algorithmic Perspective on Imitation Learning (2018, methodology): “As seen from these previous studies, advances in trajectory segmentation in imitation learning [Kulić et al., 2008, Konidaris et al., 2011, Niekum et al., 2014] are closely related to the methodological advances [Fearnhead and Liu, 2007, Fox et al., 2009] in the machine learning community.”
Skill Induction and Planning with Latent Language (2021, background): “…language strings.535 Hierarchical policies Hierarchical policy learn-536 ing and temporal abstraction have been major areas537 of focus since the earliest research on reinforce-538 ment learning and imitation learning (McGovern539 and Barto, 2001; Konidaris et al., 2012; Daniel540 et al., 2012).”
Discovering Motor Programs by Recomposing Demonstrations (2020, background): “Several works have attempted to answer this question - (Neumann et al., 2014) builds a layered approach to adapt, select, and sequence DMPs. Konidaris et al. (2012); Konidaris & Barto (2009) segments demonstrations into sequences of skills, and also merge these skills into skill-trees.”
Recent Advances in Robot Learning from Demonstration (2020, background): “Methods that learn task plans from demonstration can learn either a primitive sequence (131, 132, 140, 144) or a primitive hierarchy (127, 145–148) (see Figure 9).”
How to train your robot with deep reinforcement learning: lessons we have learned (2021, methodology): “Reinforcement learning of individual robotic skills has a long history (Daniel et al., 2013; Ijspeert et al., 2002; Kober et al., 2013; Konidaris et al., 2012; Manschitz et al., 2014; Peters et al., 2010; Peters and Schaal, 2008).”
Learning to Compose Skills (2017, methodology): “Our work is related to a family of hierarchical RL methods [4, 7, 11, 12, 10, 13].”
Learning grounded finite-state representations from unstructured demonstrations (2015, methodology): “, 2012), (inverse) reinforcement learning (Smart and Kaelbling, 2002; Abbeel and Ng, 2004; Ziebart et al., 2008; Konidaris et al., 2012), and behavior-based approaches (Nicolescu and Matarić, 2003).”
KeyIn: Discovering Subgoal Structure with Keyframe-based Video Prediction (2019, background): “L G ] 11 A p r 2019 once a subgoal sequence is discovered (Konidaris et al., 2012; Niekum et al., 2015; Kroemer et al., 2015; Hausman et al., 2017).”
A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms (2019, background): “Some approaches segment demonstration trajectories and then merge similar skills in a separate post-processing step [203], while the most principled probabilistic approaches infer shared skills across a collection of trajectories as part of the segmentation process [281, 309].”
Hierarchical visuomotor control of humanoids (2018, background): “…ing or constructing motor primitives (Ijspeert et al., 2003; Kober & Peters, 2009), deploying and sequencing them to solve tasks (Sentis & Khatib, 2005; Kober & Peters, 2014; Konidaris et al., 2012), and increasing the complexity of the control inputs to the primitives (Neumann et al., 2014).”
Learning Robot Skills with Temporal Variational Inference (2020, background): “, 2013), hierarchical approaches (Lioutikov et al., 2017; Xu et al., 2018; Huang et al., 2019; Konidaris et al., 2012; Konidaris & Barto, 2009; Shankar et al., 2020), options (Sutton et al.”
DDCO: Discovery of Deep Continuous Options for Robot Learning from Demonstrations (2017, background): “Similarly, task segmentation in Learning from Demonstrations considers partitioning a task based on demonstration data, which can be thought of as options [18, 19, 20, 21, 22, 23, 24, 8, 5, 6].”