Robot juggling: implementation of memory-based learning

By Stefan Schaal, Christopher G. Atkeson. 1994

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

Reinforcement learning in robotics: A survey (2013, background): “This approach has resulted in a variety of robot reinforcement learning applications that include pendulum swing-up tasks learned with DDP (Atkeson and Schaal, 1997; Atkeson, 1998), devil-sticking (a form of gyroscopic juggling) obtained with local LQR solutions (Schaal and Atkeson, 1994), trajector”
Reinforcement Learning: A Survey (1996, background): “Schaal and Atkeson (1994) constructed a two-armed robot, shown in Figure 11, that learns to juggle a device known as a devil-stick.”
Active Learning with Statistical Models (1996, neutral): “This contrasts with related work by Plutowski and White (1993), which is concerned with (cid:12)ltering an existing data set.”
Intrinsically Motivated Learning of Real-World Sensorimotor Skills with Developmental Constraints (2013, methodology): “Here we propose a method, inspired by the SSA algorithm (Schaal and Atkeson, 1994), to guide the system to learn on the pathway toward the selected goal position x g .”
Reinforcement Learning: An Introduction (1998, methodology): “Schaal and Atkeson (1994) applied locally weighted regression to a robot juggling control problem, where it was used to learn a system model.”
Reward Functions for Accelerated Learning (1994, methodology): “Their control systems are collections of parallel, concurrently active behaviors, some of which gather sensory information, some drive effectors, and some monitor progress and contribute reinforcement.”
Learning and Using Models (2012, methodology): “One of the earliest model-based RL methods used locally weighted regression (LWR) to learn models (Schaal and Atkeson, 1994; Atkeson et al, 1997).”
Intrinsically motivated goal exploration for active motor learning in robots: A case study (2010, methodology): “In [6], Schaal & Atkeson propose a method called SSA to deal with learning sparse data in high dimensional spaces.”
Fast, smooth and adaptive regression in metric spaces (2009, background): “[11] which considers the smooth control of a robotic tool based on noisy outside input).”
Learning to search: Functional gradient techniques for imitation learning (2009, background): “Of note is the work of Schaal and Atkeson ( Schaal and Atkeson 1994; Atkeson et al. 1995) whose algorithms develop control strategies within a given environment using trajectory demonstrations.”
Scalable Techniques from Nonparametric Statistics for Real Time Robot Learning (2002, background): “The robot uses its top two joints to perform planar devil-sticking; more details can be found in (Schaal and Atkeson, 1994).”
From model-based control to data-driven control: Survey, classification and perspective (2013, methodology): “Schaal and Atkeson first applied lazy learning algorithms to control problems in 1994 [115].”
Neural networks and differential dynamic programming for reinforcement learning problems (2016, unknown): “We consider some assumptions to simplify (A).”
Machine Learning and Radiology (2012, background): “It has broad applications in robot control (Schaal and Atkeson, 1994) and game playing (Schraudolph et al., 1994).”
A Framework for Reinforcement Learning on Real Robots (1998, background): “More recently, Schaal and Atkeson constructed a robot that successfully learned a form of juggling [ 7 ].”
Locally Weighted Learning for Control (1997, methodology): “This is done by the shifting setpoint algorithm (SSA) (Schaal and Atkeson 1994a).”