SHIV: Reducing supervisor burden in DAgger using support vectors for efficient learning from demonstrations in high dimensional state spaces

By Michael Laskey, Sam Staszak, Wesley Yu-Shu Hsieh, Jeffrey Mahler, Florian T. Pokorny, Anca Dragan, Ken Goldberg. 2016

74 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering

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

Citations in the corpus

Human-in-the-Loop Imitation Learning using Remote Teleoperation (2020, methodology): “Prior work has attempted to reduce the number of human annotations needed [10, 23, 30] but relabeling is still required.”
One-Shot Visual Imitation Learning via Meta-Learning (2017, background): “The first is the widely-studied issue of compounding errors [16, 17, 15], which we do not address in this paper.”
Query-Efficient Imitation Learning for End-to-End Simulated Driving (2017, methodology): “Instead of learning multiple policies, Laskey et al. (Laskey et al. 2016) recently introduced the SHIV approach (Svm-based reduction in Human InterVention) which converged to a single policy.”
Comparing human-centric and robot-centric sampling for robot deep learning from demonstrations (2016, background): “Similar to [13], the control space U consists of bounded changes in rotation and translation.”
Imitation Learning as f-Divergence Minimization (2019, background): “[21] Michael Laskey, Sam Staszak, Wesley Yu-Shu Hsieh, Jeffrey Mahler, Florian T Pokorny, Anca D Dragan, and Ken Goldberg.”
Mo' States Mo' Problems: Emergency Stop Mechanisms from Observation (2019, methodology): “Some recent approaches have demonstrated empirical success learning when to intervene, either in the form of resetting, collecting expert feedback, or falling back to a safe policy [8, 16, 19, 14].”
Generative Adversarial Imitation from Observation (2018, methodology): “Therefore, this method is able to learn and generalize from different demonstrations but, since it is based on behavioral cloning, it may suffer from the well-studied compounding error caused by covariate shift [24, 25, 16].”
Vision-Based Navigation With Language-Based Assistance via Imitation Learning With Indirect Intervention (2018, background): “[60, 34, 32, 37] construct policies to minimize the number of queries to the teacher.”
Active One-shot Learning (2017, background): “Applications have included domains where human or computer safety is at risk, such as cancer classification [1], malware detection [2], and autonomous driving [3, 4].”
Learning from Interventions: Human-robot interaction as both explicit and implicit feedback (2020, background): “techniques ask human experts to perform burdensome off-line labeling [5, 14, 21].”
Backplay: "Man muss immer umkehren" (2018, unknown): “These methods explicitly encourage the learned policy to mimic an expert policy (Bain & Sommut, 1999; Ross et al., 2011; Daumé et al., 2009; Zhang & Cho, 2016; Laskey et al., 2016; Nair et al., 2017; Hester et al., 2017; Ho & Ermon, 2016; Aytar et al., 2018; Lerer & Peysakhovich, 2018; Peng et al., ”
DropoutDAgger: A Bayesian Approach to Safe Imitation Learning (2017, background): “work extended DAGGER to address some inherent drawbacks [9], [10].”
Using dVRK teleoperation to facilitate deep learning of automation tasks for an industrial robot (2017, unknown): “[41] extended”