Stefan Schaal

Learning and Generalization of Motor Skills by Learning from Demonstration (2009)

Peter Pastor, Heiko Hoffmann, Tamim Asfour, Stefan Schaal

Abstract — We provide a general approach for learning robotic motor skills from human demonstration. To represent an observed movement, a non-linear differential equation is learned such that it...

Bayesian Kernel Shaping for Learning Control (2009)

Jo-anne Ting, Mrinal Kalakrishnan, Sethu Vijayakumar, Stefan Schaal

In kernel-based regression learning, optimizing each kernel individually is useful when the data density, curvature of regression surfaces (or decision boundaries) or magnitude of output noise varies...

Policy Learning for Motor Skills (2009)

Jan Peters, Stefan Schaal

Abstract. Policy learning which allows autonomous robots to adapt to novel situations has been a long standing vision of robotics, artificial intelligence, and cognitive sciences. However, to date,...

demos/digitalembryo.html (2009)

Ladan B. Shams, Mark J. Brady, Stefan Schaal

Labeled Graph Matching (LGM) has been shown successful in numerous object vision tasks. This method is the basis for arguably the best face recognition system in the world. We present an algorithm...

Segmentation of Endpoint Trajectories Does Not Imply Segmented Control (2009)

Exp Brain Res, Dagmar Sternad, Stefan Schaal

Abstract: While it is generally assumed that complex movements consist of a sequence of simpler units, the quest to define these units of action, or movement primitives, still remains an open...

Möhl, B. (eds.). Prerational Intelligence in Strategies, High-Level Processes and Collective Behavior. Kluwer Academic Press. Nonparametric Regression for Learning Nonlinear Transformations (2009)

Stefan Schaal

Information processing in animals and artificial movement systems consists of a series of transformations that map sensory signals to intermediate representations, and finally to motor commands....

Dynamic Movement Primitives–A Framework for Motor Control in Humans and Humanoid Robotics (2009)

Stefan Schaal

Given the continuous stream of movements that biological systems exhibit in their daily activities, an account for such versatility and creativity has to assume that movement sequences consist of...

DOI 10.1007/s10514-007-9051-x A unifying framework for robot control with redundant DOFs (2009)

Jan Peters, Michael Mistry, Firdaus Udwadia, Jun Nakanishi, Stefan Schaal, J. Peters, ...

2003:1783–1800, 2003) suggested to derive tracking controllers for mechanical systems with redundant degrees-offreedom (DOFs) using a generalization of Gauss ’ principle of least constraint. This...

A Bayesian Approach to Empirical Local Linearization For Robotics (2009)

Jo-anne Ting, Sethu Vijayakumar, Stefan Schaal

Abstract — Local linearizations are ubiquitous in the control of robotic systems. Analytical methods, if available, can be used to obtain the linearization, but in complex robotics systems where...

Optimization strategies in human reinforcement learning (2009)

Heiko Hoffmann, Evangelos Theodorou, Stefan Schaal

Some human movement skills require optimizing a movement such that a future event has a desired outcome. Such skills are, e.g., hitting a ball with a bat or swinging a golf club to achieve that the...

www.cc.gatech.edu/fac/Chris.Atkeson (2009)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

Abstract: Locally weighted learning (LWL) is a class of statistical learning techniques that provides useful representations and training algorithms for learning about complex phenomena during...

BOOK REVIEW The Computational Neurobiology of Reaching and Pointing — A Foundation for Motor Learning by (2009)

Reza Shadmehr, Steven P. Wise, Stefan Schaal

Over the last three decades, computational neuroscience has become an increasingly important component in neurobiological research. Modern multi-electrode and multi-site recording techniques, brain...

Learning Robot Control (2009)

Stefan Schaal

Learning robot control, a subclass of the field of learning control, refers to the process of acquiring a sensory-motor control strategy for a particular movement task and movement system by trial...

Header for SPIE use Reciprocal Excitation Between Biological and Robotic Research (2009)

Stefan Schaal, Dagmar Sternad, William Dean, Shinya Kotosaka, Rieko Osu, Mitsuo Kawato

While biological principles have inspired researchers in computational and engineering research for a long time, there is still rather limited knowledge flow back from computational to biological...

www.cc.gatech.edu/fac/Chris.Atkeson (2009)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

Abstract: Locally weighted learning (LWL) is a class of techniques from nonparametric statistics that provides useful representations and training algorithms for learning about complex phenomena...

http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained (2009)

Stefan Schaal, Auke Ijspeert, Aude Billard

prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please...

Control Systems Magazine, 14, 1, pp.57-71. Robot Juggling: An Implementation of Memory-based Learning (2009)

Stefan Schaal, Christopher G. Atkeson

This paper explores issues involved in implementing robot learning for a challenging dynamic task, using a case study from robot juggling. We use a memory-based local modeling approach (locally...

Learning an Outlier-Robust Kalman Filter (2009)

Jo-anne Ting, Evangelos Theodorou, Stefan Schaal

Abstract. We introduce a modified Kalman filter that performs robust, real-time outlier detection, without the need for manual parameter tuning by the user. Systems that rely on high quality sensory...

In: IEEE International Conference on Robotics and Automation, 3, pp.913-918, Georgia, Atlanta. Open Loop Stable Control Strategies for Robot Juggling (2009)

Stefan Schaal, Christopher G. Atkeson

Abstract: In a series of case studies out of the field of dynamic manipulation (Mason, 1992), different principles for open loop stable control are introduced and analyzed. This investigation may...

DOI 10.1007/s002210000505 RESEARCH ARTICLE (2009)

Stefan Schaal, Dagmar Sternad, S. Schaal, D. Sternad, S. Schaal, D. Sternad

Origins and violations of the 2/3 power law in rhythmic three-dimensional arm movements

Programmable pattern generators (2009)

Stefan Schaal, Dagmar Sternad

Research on perceptuomotor coordination has traditionally been split into two distinct categories, rhythmic and discrete movement. While investigations of rhythmic movement have largely been...

Information Processing Laboratories, 2-2 Hikaridai, Seika-cho, Soraku-gu, Kyoto 619-02, Japan. Receptive Field Weighted Regression (2009)

Stefan Schaal, Christopher G. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of...

A Library for Locally Weighted Projection Regression (2009)

Stefan Klanke, Sethu Vijayakumar, Stefan Schaal, Soeren Sonnenburg

In this paper we introduce an improved implementation of locally weighted projection regression (LWPR), a supervised learning algorithm that is capable of handling high-dimensional input data. As the...

Downloaded from (2009)

Jan Peters, Stefan Schaal, Stefan Schaal

One of the most general frameworks for phrasing control problems for complex, redundant robots is operational-space control. However, while this framework is of essential importance for robotics and...

On-line learning and modulation of periodic movements with nonlinear dynamical systems (2009)

Gams, Andrej, Ijspeert, Auke, Schaal, Stefan, Lenarcic, Jardan

Abstract The paper presents a two-layered system for (1) learning and encoding a periodic signal without any knowledge on its frequency and waveform, and (2) modulating the learned periodic...

Integrating Manual Dexterity with Mobility for Human-Scale Service Robotics The Case for Concentrated Research into Science and Technology Supporting Next-Generation Robotic Assistants Executive Summary 1 (2008)

Robert Ambrose, Christopher Atkeson, Oliver Brock, Rodney Brooks, Chris Brown, Joel Burdick, ...

This position paper argues that a concerted national effort to develop technologies for robotic service applications is critical and timely—targeting research on integrated systems for mobility and...

Variational Bayesian least squares: An application to brain-machine interface data (2008)

Ting, Jo-Anne, D'Souza, Aaron, Yamamoto, Kenji, Yoshioka, Toshinori, Hoffman, Donna, Kakei, Shinji, ...

An increasing number of projects in neuroscience require statistical analysis of high-dimensional data, as, for instance, in the prediction of behavior from neural firing or in the operation of...

A Bayesian Approach to Empirical Local Linearization for Robotics (2008)

Ting, Jo-Anne, D'Souza, Aaron, Vijayakumar, Sethu, Schaal, Stefan

Local linearizations are ubiquitous in the control of robotic systems. Analytical methods, if available, can be used to obtain the linearization, but in complex robotics systems where the dynamics...

Neural Computation (in press) Constructive Incremental Learning From Only Local Information (2008)

Stefan Schaal, Christopher G. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of...

WE ARE WORKING ON EASIER (2008)

Christopher G. Atkeson, Joshua G. Hale, Frank Pollick, Marcia Riley, Atr Human, Information Processing, ...

ways to program behavior in humanoid robots, and potentially in other machines and computer systems, based on how we “program” behavior in our fellow human beings. We have already demonstrated...

Summary (2008)

Humanoid Robots, Stefan Schaal

This review investigates two recent developments in artificial intelligence and neural computation: learning from imitation and the development of humanoid robots. It will be postulated that the...

Applying the Episodic Natural Actor-Critic Architecture to Motor Primitive Learning (2008)

Jan Peters, Stefan Schaal

Abstract. In this paper, we investigate motor primitive learning with the Natural Actor-Critic approach. The Natural Actor-Critic consists out of actor updates which are achieved using natural...

Predicting EMG Data from M1 Neurons with Variational Bayesian Least Squares (2008)

Jo-anne Ting, Kenji Yamamoto, Toshinori Yoshioka, Donna Hoffman, Shinji Kakei, Lauren Sergio, ...

An increasing number of projects in neuroscience requires the statistical analysis of high dimensional data sets, as, for instance, in predicting behavior from neural firing or in operating...

School of Computer and Communication Sciences, (2008)

Stefan Schaal, Auke Ijspeert, Aude Billard

Movement imitation requires a complex set of mechanisms that map an observed movement of a teacher onto one’s own movement apparatus. Relevant problems include movement recognition, pose...

LETTER Communicated by Lehel Csato Incremental Online Learning in High Dimensions (2008)

Sethu Vijayakumar, Stefan Schaal

Locally weighted projection regression (LWPR) is a new algorithm for incremental nonlinear function approximation in high-dimensional spaces with redundant and irrelevant input dimensions. At its...

Submitted to: IEEE International Conference on Humanoid Robotics Nonlinear Dynamical Systems as Movement Primitives (2008)

Stefan Schaal, Shinya Kotosaka, Dagmar Sternad

Abstract. This paper explores the idea to create complex human-like movements from movement primitives based on nonlinear attractor dynamics. Each degree-of-freedom of a limb is assumed to have two...

CONFIDENTIAL. Limited circulation. For review only. A Kalman Filter for Robust Outlier Detection (2008)

Jo-anne Ting, Evangelos Theodorou, Stefan Schaal

Abstract — In this paper, we introduce a modified Kalman filter that can perform robust, real-time outlier detection in the observations, without the need for parameter tuning. Robotic systems that...

Abstract Trajectory Formation for Imitation with Nonlinear Dynamical Systems (2008)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

This article explores a new approach to learning by imitation and trajectory formation by representing movements as mixtures of nonlinear differential equations with well-defined attractor dynamics....

MOTIVATION UNCERTAIN 3D FORCE FIELDS IN REACHING MOVEMENTS: DO HUMANS FAVOR ROBUST OR AVERAGE PERFORMANCE? (2008)

Michael Mistry, Evangelos Theodorou, Heiko Hoffman, Stefan Schaal

Scheidt et al. demonstrated that subjects exposed to a viscous force field of fixed structure but varying strength (randomly changing from trial to trial), learn to adapt to the mean disturbance,...

Bayesian Regression with Input Noise for High Dimensional Data (2008)

Jo-anne Ting, Stefan Schaal

This paper examines high dimensional regression with noise-contaminated input and output data. Goals of such learning problems include optimal prediction with noiseless query points and optimal...

2002 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2002) (to appear) A Locally Weighted Learning Composite Adaptive Controller with Structure Adaptation (2008)

Jun Nakanishi, Jay A. Farrell, Stefan Schaal

This paper introduces a provably stable adaptive learning controller which employs nonlinear function approximation with automatic growth of the learning network according to the nonlinearities and...

CONFIDENTIAL. Limited circulation. For review only. Experimental Evaluation of Task Space Position/Orientation Control Towards Compliant Control for Humanoid Robots (2008)

Jun Nakanishi, Michael Mistry, Jan Peters, Stefan Schaal

Abstract — Compliant control will be a prerequisite for humanoid robotics if these robots are supposed to work safely and robustly in human and/or dynamic environments. One view of compliant...

An Exoskeleton Robot for Human Arm Movement Study ∗ (2008)

Michael Mistry, Peyman Mohajerian, Stefan Schaal

Abstract — A new experimental platform permits us to study a novel variety of issues of human motor control, particularly full 3-D movements involving the major seven degrees-of-freedom (DOF) of...

Towards Machine Learning of Motor Skills (2008)

Jan Peters, Stefan Schaal, Bernhard Schölkopf

Abstract. Autonomous robots that can adapt to novel situations has been a long standing vision of robotics, artificial intelligence, and cognitive sciences. Early approaches to this goal during the...

Predicting EMG Data from M1 Neurons with Variational Bayesian Least Squares (2008)

Jo-anne Ting, Kenji Yamamoto, Toshinori Yoshioka, Donna Hoffman, Shinji Kakei, Lauren Sergio, ...

An increasing number of projects in neuroscience requires the statistical analysis of high dimensional data sets, as, for instance, in predicting behavior from neural firing or in operating...

Task Space Control with Prioritization for Balance and Locomotion (2008)

Michael Mistry, Jun Nakanishi, Stefan Schaal

Abstract — This paper addresses locomotion with active balancing, via task space control with prioritization. The center of gravity (COG) and foot of the swing leg are treated as task space control...

DEMONSTRATION (2008)

Christopher G. Atkeson, Stefan Schaal

The goal of robot learning from demonstration is to have a robot learn from watching a demonstration of the task to be performed. In our approach to learning from demonstration the robot learns a...

Applying the Episodic Natural Actor-Critic Architecture to Motor Primitive Learning (2008)

Jan Peters, Stefan Schaal

Abstract. In this paper, we investigate motor primitive learning with the Natural Actor-Critic approach. The Natural Actor-Critic consists out of actor updates which are achieved using natural...

www.cc.gatech.edu/fac/Chris.Atkeson (2008)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

Abstract: Locally weighted learning (LWL) is a class of techniques from nonparametric statistics that provides useful representations and training algorithms for learning about complex phenomena...

Abstract Overt Visual Attention for a Humanoid Robot (2008)

Sethu Vijayakumar, Jörg Conradt, Tomohiro Shibata, Stefan Schaal

The goal of our research is to investigate the interplay between oculomotor control, visual processing, and limb control in humans and primates by exploring the computational issues of these...

Online learning for humanoid robot systems (2008)

Jörg Conradt, Gaurav Tevatia, Sethu Vijayakumar, Stefan Schaal

Humanoid robots are high-dimensional movement systems for which analytical system identification and control methods are insufficient due to unknown nonlinearities in the system structure. As a way...

Humanoid Oculomotor Control Based on Concepts of Computational Neuroscience (2008)

Tomohiro Shibata, Jörg Conradt, Stefan Schaal

Oculomotor control in a humanoid robot faces similar problems as biological oculomotor systems, i.e., the stabilization of gaze in face of unknown perturbations of the body, selective attention, the...

WE ARE WORKING ON EASIER (2008)

Christopher G. Atkeson, Joshua G. Hale, Frank Pollick, Marcia Riley, Atr Human, Information Processing, ...

ways to program behavior in humanoid robots, and potentially in other machines and computer systems, based on how we “program” behavior in our fellow human beings. We have already demonstrated...

A Unifying Framework for Robot Control with Redundant DOFs (2008)

Peters, Jan, Mistry, Michael, Udwadia, Firdaus, Nakanishi, Jun, Schaal, Stefan

Recently, (Udwadia, 2003) suggested to derive tracking controllers for mechanical systems with redundant degrees-of-freedom (DOFs) using a generalization of Gauss’ principle of least constraint....

Bayesian Kernel Shaping for Control (2008)

Ting, Jo-Anne, Kalakrishnan, Mrinal, Vijayakumar, Sethu, Schaal, Stefan

In kernel-based regression learning, optimizing each kernel individually is useful when the data density, curvature of regression surfaces (or decision boundaries) or magnitude of output noise varies...

Generalization of Object Manipulation Skills Learned without Limb Motion (2008)

Biren Mehta, Stefan Schaal, J Neurophysiol, K. Kording, D. A. Kistemaker, ...

You might find this additional information useful... This article cites 26 articles, 8 of which you can access free at:

Online learning for humanoid robot systems (2007)

Jrg Conradt, Sethu Vijayakumar, Stefan Schaal

Humanoid robots are high-dimensional movement systems for which analytical system identification and control methods are insufficient due to unknown nonlinearities in the system structure. As a way...

An O(n) Algorithm for Real-Time Learning applied to Humanoid Robotics (2007)

Sethu Vijayakumar, Stefan Schaal

While recent research in neural networks and statistical learning has focused mostly on learning from finite data sets without stringent constraints on computational efficiency, there is an...

W E ARE WORKING ON EASIER (2007)

Christopher G. Atkeson, Joshua G. Hale, Frank Pollick, Marcia Riley, Atr Human, Information Processing, ...

ways to program behavior in humanoid robots, and potentially in other machines and computer systems, based on how we "program " behavior in our fellow human beings. We have already...

www.cc.gatech.edu/fac/Chris.Atkeson (2007)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

Abstract: Locally weighted learning (LWL) is a class of techniques from nonparametric statistics that provides useful representations and training algorithms for learning about complex phenomena...

LE?I?ER Communicated by Alexandre Potsget Constructive IncrementaI Learning from Only Local Information (2007)

Stefan Schaal, Christopher G. Atkeson

We introduce a constructive, ncremerttat learning system for regression problems that models data by means of spatally localized linear models. In contrast to other approaches, the size and shape of...

A Dynamical Approach to a Rhythmic Movement Task o (2007)

Stefan Schaal, Dagmar Sternad, Christopher G. Atkeson

The skill of rhythmic juggling a ball on a racket is investigated from the viewpoint of nonlinear dynamics. The difference equations that model the dynamical system are analyzed by means of local and...

A Library For Locally Weighted Projection Regression (2007)

Klanke, Stefan, Vijayakumar, Sethu, Schaal, Stefan

In this paper we introduce an improved implementation of locally weighted projection regression (LWPR), a supervised learning algorithm that is capable of handling high-dimensional input data. As the...

Kernel Carpentry for Online Regression using Randomly Varying Coefficient Model (2007)

Edakunni, Narayanan, Schaal, Stefan, Vijayakumar, Sethu

We present a Bayesian formulation of locally weighted learning (LWL) using the novel concept of a randomly varying coefficient model. Based on this, we propose a mechanism for multivariate non-linear...

Policy Learning for Motor Skills (2007)

Peters, Jan, Schaal, Stefan

Policy learning which allows autonomous robots to adapt to novel situations has been a long standing vision of robotics, artificial intelligence, and cognitive sciences. However, to date, learning...

Towards Machine Learning of Motor Skills (2007)

Peters, Jan, Schoelkopf, Bernhard, Schaal, Stefan

Autonomous robots that can adapt to novel situations has been a long standing vision of robotics, artificial intelligence, and cognitive sciences. Early approaches to this goal during the heydays of...

Reinforcement Learning for Optimal Control of Arm Movements (2007)

Theodorou, Evangelos, Peters, Jan, Schaal, Stefan

Every day motor behavior consists of a plethora of challenging motor skills from discrete movements such as reaching and throwing to rhythmic movements such as walking, drumming and running. How this...

Experimental evaluation of task space position/orientation control towards compliant control for humanoid robots (2007)

Nakanishi, Jun, Mistry, Michael, Peters, Jan, Schaal, Stefan

Compliant control will be a prerequisite for humanoid robotics if these robots are supposed to work safely and robustly in human and/or dynamic environments. One view of compliant control is that a...

Applying the episodic natural actor-critic architecture to motor primitive learning (2007)

Peters, Jan, Schaal, Stefan

In this paper, we investigate motor primitive learning with the Natural Actor-Critic approach. The Natural Actor-Critic consists out of actor updates which are achieved using natural stochastic...

Reinforcement learning by reward-weighted regression for operational space control (2007)

Peters, Jan, Schaal, Stefan

Many robot control problems of practical importance, including operational space control, can be reformulated as immediate reward reinforcement learning problems. However, few of the known...

Policy gradient methods for machine learning (2007)

Peters, Jan, Theodorou, Evangelos, Schaal, Stefan

We present an in-depth survey of policy gradient methods as they are used in the machine learning community for optimizing parameterized, stochastic control policies in Markovian systems with respect...

Evaluation of policy gradient methods and variants on the cart-pole benchmark (2007)

Riedmiller, Martin, Peters, Jan, Schaal, Stefan

In this paper, we evaluate different versions from the three main kinds of model-free policy gradient methods, i.e., finite difference gradients, `vanilla' policy gradients and natural policy...

Kernel carpentry for online regression using randomly varying coefficient model (2007)

Narayanan U Edakunni, Stefan Schaal, Sethu Vijayakumar

We present a Bayesian formulation of locally weighted learning (LWL) using the novel concept of a randomly varying coefficient model. Based on this, we propose a mechanism for multivariate non-linear...

Kernel carpentry for online regression using randomly varying coefficient model (2007)

Narayanan U Edakunni, Stefan Schaal, Sethu Vijayakumar

We present a Bayesian formulation of locally weighted learning (LWL) using the novel concept of a randomly varying coefficient model. Based on this, we propose a mechanism for multivariate non-linear...

Learning an Outlier-Robust Kalman Filter (2007)

Jo-anne Ting, Evangelos Theodorou, Stefan Schaal, Jo-anne Ting, Evangelos Theodorou, Stefan Schaal

Abstract. In this paper, we introduce a modified Kalman filter that performs robust, real-time outlier detection, without the need for manual parameter tuning by the user. Systems that rely on high...

Kernel carpentry for online regression using randomly varying coefficient model (2007)

Narayanan U Edakunni, Stefan Schaal, Sethu Vijayakumar

We present a Bayesian formulation of locally weighted learning (LWL) using the novel concept of a randomly varying coefficient model. Based on this, we propose a mechanism for multivariate non-linear...

Automatic outlier detection: A Bayesian approach (2007)

Jo-anne Ting, Stefan Schaal

Abstract — In order to achieve reliable autonomous control in advanced robotic systems like entertainment robots, assistive robots, humanoid robots and autonomous vehicles, sensory data needs to be...

Reinforcement learning by reward-weighted regression for operational space control (2007)

Jan Peters, Stefan Schaal

Many robot control problems of practical importance, including operational space control, can be reformulated as immediate reward reinforcement learning problems. However, few of the known...

Universal Robots—largely credited (2007)

Stefan Schaal

Research in robotics has moved away from its primary focus on industrial applications. The New Robotics is a vision that has been developed in past years by our own university and many other national...

Approximate nearest neighbor regression in very high dimensions, Nearest Neighbor Methods in Learning and Vision (2006)

Sethu Vijayakumar, Stefan Schaal

Fast and approximate nearest-neighbor search methods have recently become popular for scaling nonparameteric regression to more complex and high-dimensional applications. As an alternative to fast...

A bayesian approach to nonlinear parameter identification for rigid body dynamics (2006)

Jo-anne Ting, Michael Mistry, Jan Peters, Stefan Schaal, Jun Nakanishi

Abstract — For robots of increasing complexity such as humanoid robots, conventional identification of rigid body dynamics models based on CAD data and actuator models becomes difficult and...

Policy gradient methods for robotics (2006)

Jan Peters, Stefan Schaal

Abstract — The aquisition and improvement of motor skills and control policies for robotics from trial and error is of essential importance if robots should ever leave precisely pre-structured...

Policy gradient methods for robotics (2006)

Jan Peters, Stefan Schaal

Abstract — The aquisition and improvement of motor skills and control policies for robotics from trial and error is of essential importance if robots should ever leave precisely pre-structured...

Arm movement experiments with joint space force fields using an exoskeleton robot (2005)

Michael Mistry, Peyman Mohajerian, Stefan Schaal

Abstract — A new experimental platform permits us to study a novel variety of issues of human motor control, particularly full 3-D movements involving the major seven degrees-offreedom (DOF) of the...

Incremental Online Learning in High Dimensions (2005)

Sethu Vijayakumar, Aaron D'Souza, Stefan Schaal

Locally weighted projection regression (LWPR) is a new algorithm for incremental nonlinear function approximation in high dimensional spaces with redundant and irrelevant input dimensions. At its...

Incremental Online Learning in High Dimensions (2005)

Sethu Vijayakumar, Aaron D'Souza, Stefan Schaal

this article, however, is problematic, as it requires a careful selection of initial ridge regression parameters to stabilize the highly rank-deficient full covariance matrix of the input data, and...

Incremental Online Learning in High Dimensions (2005)

Sethu Vijayakumar, Aaron D'Souza, Stefan Schaal

Locally weighted projection regression (LWPR) is a new algorithm for incremental nonlinear function approximation in high dimensional spaces with redundant and irrelevant input dimensions. At its...

Natural actor-critic (2005)

Jan Peters, Stefan Schaal

Abstract. This paper investigates a novel model-free reinforcement learning architecture, the Natural Actor-Critic. The actor updates are based on stochastic policy gradients employing Amari’s...

A unifying methodology for the control of robotic systems (2005)

Jan Peters, Michael Mistry, Firdaus Udwadia, Jun Nakanishi, Stefan Schaal

Abstract — Recently, [1] suggested to derive tracking controllers for mechanical systems using a generalization of Gauss’ principle of least constraint. This method allows us to reformulate...

Natural actor-critic (2005)

Jan Peters, Stefan Schaal

Abstract. This paper investigates a novel model-free reinforcement learning architecture, the Natural Actor-Critic. The actor updates are based on stochastic policy gradients employing Amari’s...

Comparative experiments on task space control with redundancy resolution (2005)

Jun Nakanishi, Michael Mistry, Stefan Schaal

Abstract — Understanding the principles of motor coordination with redundant degrees of freedom still remains a challenging problem, particularly for new research in highly redundant robots like...

LWPR: A Scalable Method for Incremental Online Learning in High Dimensions (2005)

Sethu Vijayakumar, Stefan Schaal

Locally weighted projection regression (LWPR) is a new algorithm for incremental nonlinear function approximation in high dimensional spaces with redundant and irrelevant input dimensions. At its...

Predicting EMG Activity from Neural Firing in M1 with Bayesian Backfitting (2004)

Ting, Jo-Anne, D'Souza, Aaron, Schaal, Stefan

Much attention has been given to directly interpreting neural firing in the primary motor cortex as a force signal, i.e., a signal that correlates with force production in muscles. How to robustly...

Learning Motor Primitives with Reinforcement Learning (2004)

Peters, Jan, Schaal, Stefan

One of the major challenges in action generation for robotics and in the understanding of human motor control is to learn the "building blocks of move- ment generation," or more precisely, motor...

An empirical exploration of phase resetting for robust biped locomotion with dynamical movement primitives (2004)

Jun Nakanishi, Jun Morimoto, Gen Endo, Gordon Cheng, Stefan Schaal, Mitsuo Kawato

Abstract — We propose a framework for learning biped locomotion using dynamical movement primitives based on nonlinear oscillators. In our previous work, we suggested dynamical movement primitives...

the Bayesian backfitting relevance vector machine (2004)

Sethu Vijayakumar, Stefan Schaal

Traditional non-parametric statistical learning techniques are often computationally attractive, but lack the same generalization and model selection abilities as state-of-the-art Bayesian algorithms...

the Bayesian backfitting relevance vector machine (2004)

Sethu Vijayakumar, Stefan Schaal

Traditional non-parametric statistical learning techniques are often computationally attractive, but lack the same generalization and model selection abilities as state-of-the-art Bayesian algorithms...

Learning from demonstration and adaptation of biped locomotion (2004)

Jun Nakanishi, Jun Morimoto, Gen Endo, Gordon Cheng, Stefan Schaal, Mitsuo Kawato

Abstract — In this paper, we report on our research for learning biped locomotion from human demonstration. Our ultimate goal is to establish a design principle of a controller in order to achieve...

Learning composite adaptive control for a class of nonlinear systems (2004)

Jun Nakanishi, Jay A. Farrell, Stefan Schaal

Abstract — In our recent work, an adaptive composite control technique was suggested for nonlinear adaptive control with statistical learning methods. While this original work was restricted to a...

Learning movement primitives (2004)

Stefan Schaal, Jan Peters, Jun Nakanishi, Auke Ijspeert

Abstract. This paper discusses a comprehensive framework for modular motor control based on a recently developed theory of dynamic movement primitives (DMP). DMPs are a formulation of movement...

Reinforcement learning for humanoid robotics (2003)

Sethu Vijayakumar, Tomohiro Shibata, Stefan Schaal

Abstract. The complexity of the kinematic and dynamic structure of humanoid robots make conventional analytical approaches to control increasingly unsuitable for such systems. Learning techniques...

Reinforcement Learning for Humanoid Robotics (2003)

Jan Peters, Sethu Vijayakumar, Stefan Schaal

Reinforcement learning o#ers one of the most general framework to take traditional robotics towards true autonomy and versatility.

Scaling Reinforcement Learning Paradigms for Motor Control (2003)

Jan Peters, Sethu Vijayakumar, Stefan Schaal

Reinforcement learning o#ers a general framework to explain reward related learning in artificial and biological motor control. However, current reinforcement learning methods rarely scale to high...

Bayesian Backfitting (2003)

Aaron Souza Adsouza, Bayesian Backfitting, Sethu Vijayakumar, Stefan Schaal

Whenever a graphical model contains connections from multiple nodes to a single node, statistical inference of model parameters may require the evaluation and possibly the inversion of the covariance...

Reinforcement learning for humanoid robotics (2003)

Sethu Vijayakumar, Tomohiro Shibata, Stefan Schaal

Abstract. The complexity of the kinematic and dynamic structure of humanoid robots make conventional analytical approaches to control increasingly unsuitable for such systems. Learning techniques...

Reinforcement learning for humanoid robotics (2003)

Sethu Vijayakumar, Tomohiro Shibata, Stefan Schaal

Abstract. The complexity of the kinematic and dynamic structure of humanoid robots make conventional analytical approaches to control increasingly unsuitable for such systems. Learning techniques...

Learning attractor landscapes for learning motor primitives (2003)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

Many control problems take place in continuous state-action spaces, e.g., as in manipulator robotics, where the control objective is often defined as finding a desired trajectory that reaches a...

Natural Actor Critic (2003)

Jan Peters, Sethu Vijayakumar, Stefan Schaal

Reinforcement learning offers a promising framework to take planning for real-world systems towards true autonomy and versatility. However, applying reinforcement learning to high dimensional...

Movement planning and imitation by shaping nonlinear attractors (2003)

Stefan Schaal

Given the continuous stream of movements that biological systems exhibit in their daily activities, an account for such versatility and creativity has to assume that movement sequences consist of...

Movement imitation with nonlinear dynamical systems in humanoid robots (2002)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

This article presents a new approach to movement planning, on-line trajectory modification, and imitation learning by representing movement plans based on a set of nonlinear differential equations...

Learning Rhythmic Movements by Demonstration Using Nonlinear Oscillators (2002)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

This paper presents a new approach to the generation of rhythmic movement patterns with nonlinear dynamical systems. Starting from a canonical limit cycle oscillator with well-defined stability...

Learning Attractor Landscapes for Learning Motor Primitives (2002)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

Many control problems take place in continuous state-action spaces, e.g., as in manipulator robotics, where the control objective is often defined as finding a desired trajectory that reaches a...

Submitted to: Applied Intelligence Scalable Techniques from Nonparametric Statistics for Real Time Robot Learning (2002)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

Locally weighted learning (LWL) is a class of techniques from nonparametric statistics that provides useful representations and training algorithms for learning about complex phenomena during...

Biometric Oculomotor Control (2002)

Tomohiro Shibata, Sethu Vijayakumar, Jrg Conradt, Stefan Schaal

Oculomotor control in a humanoid robot faces similar problems as biological oculomotor systems, i.e., capturing targets accurately on a very narrow fovea, dealing with large delays in the control...

Learning rhythmic movements by demonstration using nonlinear oscillators (2002)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

This paper presents a new approach to the generation of rhythmic movement patterns with nonlinear dynamical systems. Starting from a canonical limit cycle oscillator with well-defined stability...

Movement imitation with nonlinear dynamical systems in humanoid robots (2002)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

This article presents a new approach to movement planning, on-line trajectory modification, and imitation learning by representing movement plans based on a set of nonlinear differential equations...

Arm and hand movement control (2002)

Stefan Schaal

The control of arm and hand movements in human and nonhuman primates has fascinated researchers in psychology, neuroscience, robotics, and numerous related areas. Movement appears effortless to the...

Forward models in visuomotor control (2002)

Biren Mehta, Stefan Schaal

Abstract: In recent years, an increasing number of research projects investigated whether the central nervous system employs internal models in motor control. While inverse models in the control loop...

Forward models in visuomotor control (2002)

Biren Mehta, Stefan Schaal

In recent years, an increasing number of research projects investigated whether the central nervous system employs internal models in motor control. While inverse models in the control loop can be...

Real-time statistical learning for robotics and human augmentation (2001)

Stefan Schaal, Sethu Vijayakumar, Auke Ijspeert, Jun Nakanishi

Abstract: Real-time modeling of complex nonlinear dynamic processes has become increasingly important in various areas of robotics and human augmentation. To address such problems, we have been...

Trajectory Formation for Imitation with Nonlinear Dynamical Systems (2001)

Auke Jan Ijspeert, Jun Nakanishi, Stefan Schaal

This article e xplore s ane approach to le rning by imitation and traje5 ory formation byre reC-- ting move - me ts as mixture s of nonline r di#e e tialeC-- tions with we ll-de fine d attractor...

Overt Visual Attention for a Humanoid Robot (2001)

Sethu Vijayakumar, Jörg Conradt, Tomohiro Shibata, Stefan Schaal

The goal of our research is to investigate the interplay between oculomotor control, visual processing, and limb control in humans and primates by exploring the computational issues of these...

Learning Inverse Kinematics (2001)

Aaron D'Souza, Sethu Vijayakumar, Stefan Schaal

Real-time control of the ende#ector of a humanoid robot in external coordinates requires computationally e#cient solutions of the inverse kinematics problem. In this context, this paper investigates...

Real-Time Statistical Learning For Robotics and (2001)

Human Augmentation Stefan, Stefan Schaal, Sethu Vijayakumar, Auke Ijspeert, Jun Nakanishi

Real-time modeling of complex nonlinear dynamic processes has become increasingly important in various areas of robotics and human augmentation. To address such problems, we have been developing...

Learning inverse kinematics (2001)

Sethu Vijayakumar, Stefan Schaal

Real-time control of the endeffector of a humanoid robot in external coordinates requires computationally efficient solutions of the inverse kinematics problem. In this context, this paper...

Nonlinear dynamical systems for imitation with humanoid robots (2001)

Auke Jan Ijspeert, Jun Nakanishi, Tomohiro Shibata, Stefan Schaal

This article explores a new approach to learning by imitation and trajectory formation by representing movements as control policies (CPs) based on a set of nonlinear differential equations with...

Real-time statistical learning for robotics and human augmentation (2001)

Stefan Schaal, Sethu Vijayakumar, Auke Ijspeert, Jun Nakanishi

Abstract: Real-time modeling of complex nonlinear dynamic processes has become increasingly important in various areas of robotics and human augmentation. To address such problems, we have been...

Origins and violations of the 2/3 power law in rhythmic 3D movements (2001)

Stefan Schaal, Dagmar Sternad

The 2/3 power law, the nonlinear relationship between tangential velocity and radius of curvature of the endeffector trajectory, has been suggested as a fundamental constraint of the central nervous...

Fast and efficient incremental learning for high-dimensional movement systems (2000)

Sethu Vijayakumar, Stefan Schaal

Abstract: We introduce a new algorithm, Locally Weighted Projection Regression (LWPR), for incremental real-time learning of nonlinear functions, as particularly useful for problems of autonomous...

Locally Weighted Projection Regression : An O(n) algorithm for incremental real time learning in high dimensional space (2000)

Sethu Vijayakumar, Stefan Schaal

Locally weighted projection regression is a new algorithm that achieves nonlinear function approximation in high dimensional spaces with redundant and irrelevant input dimensions. At its core, it...

Real-time robot learning with locally weighted statistical learning (2000)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

Abstract: Locally weighted learning (LWL) is a class of statistical learning techniques that provides useful representations and training algorithms for learning about complex phenomena during...

Real-Time Robot Learning With Locally Weighted Statistical Learning (2000)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

: Locally weighted learning (LWL) is a class of statistical learning techniques that provides useful representations and training algorithms for learning about complex phenomena during autonomous...

Real Time Learning in Humanoids: A Challenge for Scalability of Online Algorithms (2000)

Sethu Vijayakumar, Stefan Schaal

While recent research in neural networks and statistical learning has focused mostly on learning from finite data sets without stringent constraints on computational efficiency, there is an...

Locally weighted projection regression: An o(n) algorithm for incremental real time learning in high dimensional space (2000)

Sethu Vijayakumar, Stefan Schaal

Locally weighted projection regression is a new algorithm that achieves nonlinear function approximation in high dimensional spaces with redundant and irrelevant input dimensions. At its core, it...

Real-time robot learning with locally weighted statistical learning (2000)

Stefan Schaal, Christopher G. Atkeson, Sethu Vijayakumar

Abstract: Locally weighted learning (LWL) is a class of statistical learning techniques that provides useful representations and training algorithms for learning about complex phenomena during...

Fast and efficient incremental learning for high-dimensional movement systems (2000)

Sethu Vijayakumar, Stefan Schaal

Abstract: We introduce a new algorithm, Locally Weighted Projection Regression (LWPR), for incremental real-time learning of nonlinear functions, as particularly useful for problems of autonomous...

Inverse kinematics for humanoid robots (2000)

Gaurav Tevatia, Stefan Schaal

Abstract: Real-time control of the endeffector of a humanoid robot in external coordinates requires computationally efficient solutions of the inverse kinematics problem. In this context, this paper...

Using humanoid robots to study human behavior (2000)

Christopher G. Atkeson, Joshua G. Hale, Frank Pollick, Marcia Riley, Atr Human, Information Processing, ...

ways to program behavior in humanoid robots, and potentially in other machines and computer systems, based on how we “program” behavior in our fellow human beings. We have already demonstrated...

Using humanoid robots to study human behavior (2000)

Christopher G. Atkeson, Mitsuo Kawato, Shinya Kotosaka, Frank Pollick, Marcia Riley, Stefan Schaal, ...

www.erato.atr.co.jp/DB/ We are using humanoid robots to explore computational models of how human behavior is generated. Using a humanoid robot as a research tool forces us to deal with a complex...

Biomimetic Gaze Stabilization (1999)

Tomohiro Shibata, Stefan Schaal

Accurate oculomotor control is one of the essential pre-requisites for successful visuomotor coordination. In this paper, we suggest a biologically inspired control system for learning gaze...

Biomimetic Gaze Stabilization (1999)

Tomohiro Shibata, Stefan Schaal

Accurate oculomotor control is one of the essential pre-requisites for successful visuomotor coordination. In this paper, we suggest a biologically inspired control system for learning gaze...

Programmable Pattern Generators (1998)

Stefan Schaal, Dagmar Sternad

: This paper explores the idea to create complex human-like arm movements from movement primitives based on nonlinear attractor dynamics. Each degree-offreedom of an arm is assumed to have two...

Local Adaptive Subspace Regression (1998)

Sethu Vijayakumar, Stefan Schaal

Incremental learning of sensorimotor transformations in high dimensional spaces is one of the basic prerequisites for the success of autonomous robot devices as well as biological movement systems....

Local dimensionality reduction (1998)

Stefan Schaal, Sethu Vijayakumar, Christopher G. Atkeson

If globally high dimensional data has locally only low dimensional distributions, it is advantageous to perform a local dimensionality reduction before further processing the data. In this paper we...

Local adaptive subspace regression (1998)

Sethu Vijayakumar, Stefan Schaal

Abstract. Incremental learning of sensorimotor transformations in high dimensional spaces is one of the basic prerequisites for the success of autonomous robot devices as well as biological movement...

Local adaptive subspace regression (1998)

Sethu Vijayakumar, Stefan Schaal

Abstract: Incremental learning of sensorimotor transformations in high dimensional spaces is one of the basic prerequisites for the success of autonomous robot devices as well as biological movement...

Constructive incremental learning from only local information (1998)

Stefan Schaal, Christopher G. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of...

Robot learning from demonstration (1997)

Stefan Schaal

By now it is widely accepted that learning a task from scratch, i.e., without any prior knowledge, is a daunting undertaking. Humans, however, rarely attempt to learn from scratch. They extract...

Constructive Incremental Learning From Only Local Information (1997)

Stefan Schaal, Christopher G. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of...

Learning from Demonstration (1997)

Stefan Schaal

By now it is widely accepted that learning a task from scratch, i.e., without any prior knowledge, is a daunting undertaking. Humans, however, rarely attempt to learn from scratch. They extract...

Local Dimensionality Reduction (1997)

Stefan Schaal, Sethu Vijayakumar, Christopher C. Atkeson

If globally high dimensional data has locally only low dimensional distributions, it is advantageous to perform a local dimensionality reduction before further processing the data. In this paper we...

Robot Learning From Demonstration (1997)

Christopher G. Atkeson, Stefan Schaal

The goal of robot learning from demonstration is to have a robot learn from watching a demonstration of the task to be performed. In our approach to learning from demonstration the robot learns a...

Local Dimensionality Reduction for Locally Weighted Learning (1997)

Sethu Vijayakumar, Stefan Schaal

Incremental learning of sensorimotor transformations in high dimensional spaces is one of the basic prerequisites for the success of autonomous robot devices as well as biological movement systems....

Receptive Field Weighted Regression (1997)

Stefan Schaal, C. G. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of...

Learning from demonstration (1997)

Stefan Schaal

By now it is widely accepted that learning a task from scratch, i.e., without any prior knowledge, is a daunting undertaking. Humans, however, rarely attempt to learn from scratch. They extract...

One-handed juggling: A dynamical approach to a rhythmic movement task (1996)

Stefan Schaal, Dagmar Sternad, Christopher G. Atkeson

The skill of rhythmic juggling a ball on a racket is investigated from the viewpoint of nonlinear dynamics. The difference equations that model the dynamical system are analyzed by means of local and...

From Isolation to Cooperation: An Alternative View of a System of Experts (1996)

Stefan Schaal, C.C. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of locally linear experts. In contrast to other approaches, the experts are trained...

Locally Weighted Learning (1996)

Christopher G. Atkeson, Andrew W. Moore, Stefan Schaal

This paper surveys locally weighted learning, a form of lazy learning and memory based learning, and focuses on locally weighted linear regression. The survey discusses distance functions, smoothing...

Locally Weighted Learning for Control (1996)

Christopher G. Atkeson, Andrew W. Moore, Stefan Schaal

Lazy learning methods provide useful representations and training algorithms for learning about complex phenomena during autonomous adaptive control of complex systems. This paper surveys ways in...

From isolation to cooperation: An alternative view of a system of experts (1996)

Stefan Schaal, Christopher C. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of locally linear experts. In contrast to other approaches, the experts are trained...

One-handed juggling: A dynamical approach to a rhythmic movement task (1996)

Stefan Schaal, Dagmar Sternad, Christopher G. Atkeson

The skill of rhythmic juggling a ball on a racket is investigated from the viewpoint of nonlinear dynamics. The difference equations that model the dynamical system are analyzed by means of local and...

Robot learning by nonparametric regression (1995)

Stefan Schaal, Christopher G. Atkeson

Abstract: We present an approach to robot learning grounded on a nonparametric regression technique, locally weighted regression. The model of the task to be performed is represented by infinitely...

From Isolation to Cooperation: An Alternative View of a System of Experts (1995)

Stefan Schaal, Christopher C. Atkeson

: We introduce a constructive, incremental learning system for regression problems that models data by means of locally linear experts. In contrast to other approaches, the experts are trained...

Memory-Based Neural Networks For Robot Learning (1995)

Christopher G. Atkeson, Stefan Schaal

This paper explores a memory-based approach to robot learning, using memorybased neural networks to learn models of the task to be performed. Steinbuch and Taylor presented neural network designs to...

Memory-based neural networks for robot learning (1995)

Christopher G. Atkeson, Stefan Schaal

This paper explores a memory-based approach to robot learning, using memorybased neural networks to learn models of the task to be performed. Steinbuch and Taylor presented neural network designs to...

Assessing the quality of learned local models (1994)

Stefan Schaal, Christopher G. Atkeson

An approach is presented to learning high dimensional functions in the case where the learning algorithm can affect the generation of new data. A local modeling algorithm, locally weighted...

Nonparametric Regression for Learning (1994)

Stefan Schaal

: In recent years, learning theory has been increasingly influenced by the fact that many learning algorithms have at least in part a comprehensive interpretation in terms of well established...

Assessing the quality of learned local models (1994)

Stefan Schaal, Christopher G. Atkeson

An approach is presented to learning high dimensional functions in the case where the learning algorithm can affect the generation of new data. A local modeling algorithm, locally weighted...

Nonparametric regression for learning (1994)

Stefan Schaal

Abstract: In recent years, learning theory has been increasingly influenced by the fact that many learning algorithms have at least in part a comprehensive interpretation in terms of well established...

Robot learning by nonparametric regression (1994)

Stefan Schaal, Christopher G. Atkeson

Abstract: We present an approach to robot learning grounded on a nonparametric regression technique, locally weighted regression. The model of the task to be performed is represented by infinitely...

Learning passive motor control strategies with genetic algorithms (1993)

Stefan Schaal, Dagmar Sternad

Abstract: This study investigates learning passive motor control strategies. Passive control is understood as control without active error correction; the movement is stabilized by particular...

Constructive Incremental Learning From Only Local Information (1992)

Stefan Schaal, Christopher G. Atkeson

We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of...

Computational approaches to motor learning by imitation.

Schaal, Stefan, Ijspeert, Auke, Billard, Aude

Movement imitation requires a complex set of mechanisms that map an observed movement of a teacher onto one's own movement apparatus. Relevant problems include movement recognition, pose estimation,...

The New Robotics—towards human-centered machines

Schaal, Stefan

Research in robotics has moved away from its primary focus on industrial applications. The New Robotics is a vision that has been developed in past years by our own university and many other national...