Computational approaches to motor learning by imitation.

By Stefan Schaal, Auke Jan Ijspeert, Aude Gemma Billard. 2003

2,692 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Control and Systems Engineering, Social Psychology

How researchers see this paper

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

In Cognitive Neuroscience

In Control and Systems Engineering

In Social Psychology

Citations in the corpus

Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations (2017, neutral): “c) Imitation learning: In imitation learning, demonstrations of successful behavior are used to train policies that imitate the expert providing these successful trajectories [42].”
One-Shot Visual Imitation Learning via Meta-Learning (2017, positive): “We present a method that combines imitation learning [6] with meta-learning [7] for one-shot learning from visual demonstrations.”
One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning (2018, neutral): “Most imitation learning and learning from demonstration methods operate at the level of configuration-space trajectories [44, 2], which are typically collected using kinesthetic teaching [32, 1], teleoperation [8, 35, 62], or sensors on the”
Coresets for Scalable Bayesian Logistic Regression (2016, methodology): “The analysis in this work allows any prior π0(θ); common choices are the Gaussian, Cauchy [12], and spike-and-slab [13].”
BM: An iterative algorithm to learn stable non-linear dynamical systems with Gaussian mixture models (2010, neutral): “Dynamical systems (DS) have been recently advocated as a powerful means of modeling robot motions [5]–[7].”
Handling Sparsity via the Horseshoe (2009, methodology): “The first approach (Mitchell and Beauchamp, 1988; George and McCulloch, 1993) models each β i with a prior comprising both a point mass at β i = 0 and an absolutely continuous alternative; the second approach (see, e.g., Tibshirani, 1996 and Tipping , 2001) models the β i 's with absolutely…”
Inverse groundwater modeling for hydraulic conductivity estimation using Bayesian model averaging and variance window (2008, methodology): “The Markov chain Monte Carlo (MCMC) method was suggested to calculate the integration [George and McCulloch, 1993; Madigan and Raftery, 1994; Madigan and York, 1995; Raftery et al., 1997].”
A survey of Bayesian predictive methods for model assessment, selection and comparison (2012, background): “…the parameters to specific values, the BMA predictive distribution over all the submodels can be equivalently formulated by placing a discrete prior probability for the said specific values and integrating over the parameters (e.g., George and McCulloch, 1993; Brown, Vannucci and Fearn, 1998).”
Variable selection and sensitivity analysis using dynamic trees, with an application to computer code performance tuning (2011, methodology): “Stochastic-search variable selection [SSVS, George and McCulloch (1993)] is used for selecting main eﬀects and interactions.”
Polynomial nonlinear spatio‐temporal integro‐difference equation models (2011, methodology): “George and McCulloch (1993, 1997) provide various alternatives for the specification of these hyperparameters.”
A Unified Framework for Fitting Bayesian Semiparametric Models to Arbitrarily Censored Survival Data, Including Spatially Referenced Data (2017, methodology): “ posterior probabilities, i.e. a ranking of models much like the best subsets C p statistic. This variable selection method was originally termed \stochastic search variable selection" (SSVS) by George and McCulloch (1993) who instead of using Bernoulli point masses for each regression eect use”
Adaptive Shrinkage in Bayesian Vector Autoregressive Models (2019, background): “For instance, George, Sun, and Ni (2008) proposed stochastic search variable selection (SSVS) priors (see George and McCulloch 1993) that impose a mixture of Gaussians prior on each regression coefficient in the VAR.”
Objective Testing Procedures in Linear Models: Calibration of the p‐values (2006, background): “Some recent key papers are George & McCulloch (1993, 1997), George & Foster (2000), Chipman et al. (2001) and Casella & Moreno (2006).”
Empirical Bayes vs. Fully Bayes Variable Selection (2008, background): “As the default of w a ,w b , we consider w a = w b = 1, which yields 1,1 () = 1 p + 1 p q −1 , (14) a special case of a form proposed by George and McCulloch (1993).”
Big Data and Data Science Methods for Management Research (2016, background): “To this end, George and McCulloch (1993) introduce Bayesian regression with a latent variable, 𝛾𝑗, that indicates whether variable j is selected.”
Bayesian functional ANOVA modeling using Gaussian process prior distributions (2010, background): “…comparison of sub-models in which entire factors or sets of interactions are present or absent, one may consider introducing latent indicator variables for the event that a particular variance component is identically zero, an idea following from the seminal paper of George and McCulloch (1993).”
An integrated hydrologic Bayesian multimodel combination framework: Confronting input, parameter, and model structural uncertainty in hydrologic prediction (2006, methodology): “In many case studies, the BMA has shown to produce more accurate and reliable predictions than other multimodel techniques [George and McCulloch, 1993; Raftery et al., 1997; Clyde, 1999; Viallefont et al., 2001; Raftery and Zheng, 2003; Ellison, 2004].”