Composing Complex Skills by Learning Transition Policies

By Youngwoon Lee, Shao-Hua Sun, Sriram Somasundaram, Edward S. Hu, Joseph J. Lim. 2018

103 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.

Citations in the corpus

Accelerating Reinforcement Learning with Learned Skill Priors (2020, background): “Learned skills can be represented as sub-policies in the form of options [33, 1], as subgoal setter and reacher functions [9, 21] or discrete primitive libraries [27, 17].”
Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation (2019, background): “While impressive results have been shown, it is nontrivial to adopt them for complex tasks such as acquiring robotic skills using reinforcement learning [12, 22, 14, 33, 9, 10, 20].”
IKEA Furniture Assembly Environment for Long-Horizon Complex Manipulation Tasks (2019, background): “Recently, many simulated environments have been introduced in diverse applications, such as autonomous driving [12, 13], indoor navigation [14, 15], continuous control [16–18], and recommendation systems [19].”
Learning to Synthesize Programs as Interpretable and Generalizable Policies (2021, background): “[82] Youngwoon Lee, Shao-Hua Sun, Sriram Somasundaram, Edward Hu, and Joseph J.”
Skill-based Meta-Reinforcement Learning (2022, background): “Methods for skill-based learning recombine these skills for learning unseen target tasks and converge substantially faster than methods that learn from scratch (Lee et al., 2018; Hausman et al., 2018; Sharma et al., 2020; Sun, 2022).”
Learning to Coordinate Manipulation Skills via Skill Behavior Diversification (2020, background): “…(Schmidhuber, 1990; Bacon et al., 2017; Nachum et al., 2018; Levy et al., 2019), meta-learned (Frans et al., 2018), pre-deﬁned (Kulkarni et al., 2016; Oh et al., 2017; Merel et al., 2019; Lee et al., 2019), or attained from additional supervision signals (Andreas et al., 2017; Ghosh et al., 2018).”
Hierarchical Style-based Networks for Motion Synthesis (2020, background): “This has not been deeply addressed in previous motion completion works [33,16].”
How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Composition (2022, background): “For example, the approach of Lee et al. (2019) learns a policy for transitioning from one skill to the next, given a set of pretrained skills.”
CompILE: Compositional Imitation Learning and Execution (2018, background): “, 2017) and learning of transition policies (Lee et al., 2019) has also been addressed in this context.”