The Experts below are selected from a list of 4833 Experts worldwide ranked by ideXlab platform
Pierre-yves Oudeyer - One of the best experts on this subject based on the ideXlab platform.
-
Developmental Learning for Intelligent Tutoring Systems
4th International Conference on Development and Learning and on Epigenetic Robotics, 2014Co-Authors: Benjamin Clement, Didier Roy, Pierre-yves Oudeyer, Manuel LopesAbstract:We present an approach to Intelligent Tutoring Systems which adaptively personalizes sequences of Learning activities to maximize skills acquired by students, taking into account the limited time and motivational resources. At a given point in time, the system tries to propose to the student the activity which makes him Progress best. We introduce two algorithms that rely on the empirical estimation of the Learning Progress, one that uses information about the difficulty of each exercise RiARiT and another that does not use any knowledge about the problem ZPDES. The system is based on the combination of three approaches. First, it leverages recent models of intrinsically motivated Learning by transposing them to active teaching, relying on empirical estimation of Learning Progress provided by specific activities to particular students. Second, it uses state-of-the-art Multi-Arm Bandit (MAB) techniques to efficiently manage the exploration/exploitation challenge of this optimization process. Third, it leverages expert knowledge to constrain and bootstrap initial exploration of the MAB, while requiring only coarse guidance information of the expert and allowing the system to deal with didactic gaps in its knowledge.
-
Online Optimization of \mbox{Teaching} Sequences with Multi-Armed Bandits
Proceedings of the 7th International Conference on Educational Data Mining, 2014Co-Authors: Benjamin Clement, Didier Roy, Pierre-yves OudeyerAbstract:We present an approach to Intelligent Tutoring Systems which adaptively personalizes sequences of Learning activi- ties to maximize skills acquired by each student, taking into account limited time and motivational resources. At a given point in time, the system tries to propose to the student the activity which makes him Progress best. We introduce two algorithms that rely on the empirical estimation of the Learning Progress, one that uses information about the dif- ficulty of each exercise RiARiT and another that does not use any knowledge about the problem ZPDES. The system is based on the combination of three approaches. First, it leverages recent models of intrinsically motivated Learning by transposing them to active teaching, relying on empirical estimation of Learning Progress provided by spe- cific activities to particular students. Second, it uses state- of-the-artMulti-Arm Bandit (MAB) techniques to efficiently manage the exploration/exploitation challenge of this op- timization process. Third, it leverages expert knowledge to constrain and bootstrap initial exploration of the MAB, while requiring only coarse guidance information of the ex- pert and allowing the system to deal with didactic gaps in its knowledge.
-
Multi-Armed Bandits for Intelligent Tutoring Systems
CoRR, 2013Co-Authors: Manuel Lopes, Didier Roy, Benjamin Clement, Pierre-yves OudeyerAbstract:We present an approach to Intelligent Tutoring Systems which adaptively personalizes sequences of Learning activities to maximize skills acquired by students, taking into account the limited time and motivational resources. At a given point in time, the system proposes to the students the activity which makes them Progress faster. We introduce two algorithms that rely on the empirical estimation of the Learning Progress, RiARiT that uses information about the difficulty of each exercise and ZPDES that uses much less knowledge about the problem. The system is based on the combination of three approaches. First, it leverages recent models of intrinsically motivated Learning by transposing them to active teaching, relying on empirical estimation of Learning Progress provided by specific activities to particular students. Second, it uses state-of-the-art Multi-Arm Bandit (MAB) techniques to efficiently manage the exploration/exploitation challenge of this optimization process. Third, it leverages expert knowledge to constrain and bootstrap initial exploration of the MAB, while requiring only coarse guidance information of the expert and allowing the system to deal with didactic gaps in its knowledge. The system is evaluated in a scenario where 7-8 year old schoolchildren learn how to decompose numbers while manipulating money. Systematic experiments are presented with simulated students, followed by results of a user study across a population of 400 school children.
-
exploration in model based reinforcement Learning by empirically estimating Learning Progress
Neural Information Processing Systems, 2012Co-Authors: Manuel Lopes, Tobias Lang, Marc Toussaint, Pierre-yves OudeyerAbstract:Formal exploration approaches in model-based reinforcement Learning estimate the accuracy of the currently learned model without consideration of the empirical prediction error. For example, PAC-MDP approaches such as R-MAX base their model certainty on the amount of collected data, while Bayesian approaches assume a prior over the transition dynamics. We propose extensions to such approaches which drive exploration solely based on empirical estimates of the learner's accuracy and Learning Progress. We provide a "sanity check" theoretical analysis, discussing the behavior of our extensions in the standard stationary finite state-action case. We then provide experimental studies demonstrating the robustness of these exploration measures in cases of non-stationary environments or where original approaches are misled by wrong domain assumptions.
-
NIPS - Exploration in Model-based Reinforcement Learning by Empirically Estimating Learning Progress
2012Co-Authors: Manuel Lopes, Tobias Lang, Marc Toussaint, Pierre-yves OudeyerAbstract:Formal exploration approaches in model-based reinforcement Learning estimate the accuracy of the currently learned model without consideration of the empirical prediction error. For example, PAC-MDP approaches such as R-MAX base their model certainty on the amount of collected data, while Bayesian approaches assume a prior over the transition dynamics. We propose extensions to such approaches which drive exploration solely based on empirical estimates of the learner's accuracy and Learning Progress. We provide a "sanity check" theoretical analysis, discussing the behavior of our extensions in the standard stationary finite state-action case. We then provide experimental studies demonstrating the robustness of these exploration measures in cases of non-stationary environments or where original approaches are misled by wrong domain assumptions.
Atsuo Takanishi - One of the best experts on this subject based on the ideXlab platform.
-
development of a suture ligature training system designed to provide quantitative information of the Learning Progress of trainees
International Conference on Robotics and Automation, 2007Co-Authors: N Oshima, M Aizudding, R Midorikawa, Jorge Solis, Yu Ogura, Atsuo TakanishiAbstract:Surgeons, during a medical intervention, perform different kinds of manual tasks with dexterity and precision. In order to assure the success of the intervention, medical students are trained several years using training models so that they can perform such tasks with accuracy to avoid any possible risk to patients. However, current training models are merely designed to imitate the surgical procedure without providing any further information about the how well the task was done. For that reason; in this paper, the development of a suture/ligature training system is proposed to imitate surgical procedures as well as provide quantitative information of the Learning Progress of trainees. As a first approach, a training system was designed to simulate the suture and ligature tasks. The proposed training system includes a skin dummy with an array of embedded photo interrupters to detect the movement of the skin dummy; without requiring any modification on the surgical instrument. In this paper, different task parameters were proposed to understand trainees' Learning Progress and different experiments were carried out to identify the evaluation parameters that may provide useful information about how well the task was performed. As a result from the experiments, the evaluation parameters for the suture and ligature were determined. Finally, an evaluation function was proposed and further experiments were proposed to verify its effectiveness. From the results of the experiments, we could effectively distinguish quantitatively the differences of skill levels between surgeons and unskilled persons as well as identifying the Learning Progress of trainees by plotting the Learning curve
-
ICRA - Development of a Suture/Ligature Training System designed to provide quantitative information of the Learning Progress of trainees
Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007Co-Authors: N Oshima, M Aizudding, R Midorikawa, Jorge Solis, Yu Ogura, Atsuo TakanishiAbstract:Surgeons, during a medical intervention, perform different kinds of manual tasks with dexterity and precision. In order to assure the success of the intervention, medical students are trained several years using training models so that they can perform such tasks with accuracy to avoid any possible risk to patients. However, current training models are merely designed to imitate the surgical procedure without providing any further information about the how well the task was done. For that reason; in this paper, the development of a suture/ligature training system is proposed to imitate surgical procedures as well as provide quantitative information of the Learning Progress of trainees. As a first approach, a training system was designed to simulate the suture and ligature tasks. The proposed training system includes a skin dummy with an array of embedded photo interrupters to detect the movement of the skin dummy; without requiring any modification on the surgical instrument. In this paper, different task parameters were proposed to understand trainees' Learning Progress and different experiments were carried out to identify the evaluation parameters that may provide useful information about how well the task was performed. As a result from the experiments, the evaluation parameters for the suture and ligature were determined. Finally, an evaluation function was proposed and further experiments were proposed to verify its effectiveness. From the results of the experiments, we could effectively distinguish quantitatively the differences of skill levels between surgeons and unskilled persons as well as identifying the Learning Progress of trainees by plotting the Learning curve
N Oshima - One of the best experts on this subject based on the ideXlab platform.
-
development of a suture ligature training system designed to provide quantitative information of the Learning Progress of trainees
International Conference on Robotics and Automation, 2007Co-Authors: N Oshima, M Aizudding, R Midorikawa, Jorge Solis, Yu Ogura, Atsuo TakanishiAbstract:Surgeons, during a medical intervention, perform different kinds of manual tasks with dexterity and precision. In order to assure the success of the intervention, medical students are trained several years using training models so that they can perform such tasks with accuracy to avoid any possible risk to patients. However, current training models are merely designed to imitate the surgical procedure without providing any further information about the how well the task was done. For that reason; in this paper, the development of a suture/ligature training system is proposed to imitate surgical procedures as well as provide quantitative information of the Learning Progress of trainees. As a first approach, a training system was designed to simulate the suture and ligature tasks. The proposed training system includes a skin dummy with an array of embedded photo interrupters to detect the movement of the skin dummy; without requiring any modification on the surgical instrument. In this paper, different task parameters were proposed to understand trainees' Learning Progress and different experiments were carried out to identify the evaluation parameters that may provide useful information about how well the task was performed. As a result from the experiments, the evaluation parameters for the suture and ligature were determined. Finally, an evaluation function was proposed and further experiments were proposed to verify its effectiveness. From the results of the experiments, we could effectively distinguish quantitatively the differences of skill levels between surgeons and unskilled persons as well as identifying the Learning Progress of trainees by plotting the Learning curve
-
ICRA - Development of a Suture/Ligature Training System designed to provide quantitative information of the Learning Progress of trainees
Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007Co-Authors: N Oshima, M Aizudding, R Midorikawa, Jorge Solis, Yu Ogura, Atsuo TakanishiAbstract:Surgeons, during a medical intervention, perform different kinds of manual tasks with dexterity and precision. In order to assure the success of the intervention, medical students are trained several years using training models so that they can perform such tasks with accuracy to avoid any possible risk to patients. However, current training models are merely designed to imitate the surgical procedure without providing any further information about the how well the task was done. For that reason; in this paper, the development of a suture/ligature training system is proposed to imitate surgical procedures as well as provide quantitative information of the Learning Progress of trainees. As a first approach, a training system was designed to simulate the suture and ligature tasks. The proposed training system includes a skin dummy with an array of embedded photo interrupters to detect the movement of the skin dummy; without requiring any modification on the surgical instrument. In this paper, different task parameters were proposed to understand trainees' Learning Progress and different experiments were carried out to identify the evaluation parameters that may provide useful information about how well the task was performed. As a result from the experiments, the evaluation parameters for the suture and ligature were determined. Finally, an evaluation function was proposed and further experiments were proposed to verify its effectiveness. From the results of the experiments, we could effectively distinguish quantitatively the differences of skill levels between surgeons and unskilled persons as well as identifying the Learning Progress of trainees by plotting the Learning curve
Manuel Lopes - One of the best experts on this subject based on the ideXlab platform.
-
Developmental Learning for Intelligent Tutoring Systems
4th International Conference on Development and Learning and on Epigenetic Robotics, 2014Co-Authors: Benjamin Clement, Didier Roy, Pierre-yves Oudeyer, Manuel LopesAbstract:We present an approach to Intelligent Tutoring Systems which adaptively personalizes sequences of Learning activities to maximize skills acquired by students, taking into account the limited time and motivational resources. At a given point in time, the system tries to propose to the student the activity which makes him Progress best. We introduce two algorithms that rely on the empirical estimation of the Learning Progress, one that uses information about the difficulty of each exercise RiARiT and another that does not use any knowledge about the problem ZPDES. The system is based on the combination of three approaches. First, it leverages recent models of intrinsically motivated Learning by transposing them to active teaching, relying on empirical estimation of Learning Progress provided by specific activities to particular students. Second, it uses state-of-the-art Multi-Arm Bandit (MAB) techniques to efficiently manage the exploration/exploitation challenge of this optimization process. Third, it leverages expert knowledge to constrain and bootstrap initial exploration of the MAB, while requiring only coarse guidance information of the expert and allowing the system to deal with didactic gaps in its knowledge.
-
Multi-Armed Bandits for Intelligent Tutoring Systems
CoRR, 2013Co-Authors: Manuel Lopes, Didier Roy, Benjamin Clement, Pierre-yves OudeyerAbstract:We present an approach to Intelligent Tutoring Systems which adaptively personalizes sequences of Learning activities to maximize skills acquired by students, taking into account the limited time and motivational resources. At a given point in time, the system proposes to the students the activity which makes them Progress faster. We introduce two algorithms that rely on the empirical estimation of the Learning Progress, RiARiT that uses information about the difficulty of each exercise and ZPDES that uses much less knowledge about the problem. The system is based on the combination of three approaches. First, it leverages recent models of intrinsically motivated Learning by transposing them to active teaching, relying on empirical estimation of Learning Progress provided by specific activities to particular students. Second, it uses state-of-the-art Multi-Arm Bandit (MAB) techniques to efficiently manage the exploration/exploitation challenge of this optimization process. Third, it leverages expert knowledge to constrain and bootstrap initial exploration of the MAB, while requiring only coarse guidance information of the expert and allowing the system to deal with didactic gaps in its knowledge. The system is evaluated in a scenario where 7-8 year old schoolchildren learn how to decompose numbers while manipulating money. Systematic experiments are presented with simulated students, followed by results of a user study across a population of 400 school children.
-
exploration in model based reinforcement Learning by empirically estimating Learning Progress
Neural Information Processing Systems, 2012Co-Authors: Manuel Lopes, Tobias Lang, Marc Toussaint, Pierre-yves OudeyerAbstract:Formal exploration approaches in model-based reinforcement Learning estimate the accuracy of the currently learned model without consideration of the empirical prediction error. For example, PAC-MDP approaches such as R-MAX base their model certainty on the amount of collected data, while Bayesian approaches assume a prior over the transition dynamics. We propose extensions to such approaches which drive exploration solely based on empirical estimates of the learner's accuracy and Learning Progress. We provide a "sanity check" theoretical analysis, discussing the behavior of our extensions in the standard stationary finite state-action case. We then provide experimental studies demonstrating the robustness of these exploration measures in cases of non-stationary environments or where original approaches are misled by wrong domain assumptions.
-
NIPS - Exploration in Model-based Reinforcement Learning by Empirically Estimating Learning Progress
2012Co-Authors: Manuel Lopes, Tobias Lang, Marc Toussaint, Pierre-yves OudeyerAbstract:Formal exploration approaches in model-based reinforcement Learning estimate the accuracy of the currently learned model without consideration of the empirical prediction error. For example, PAC-MDP approaches such as R-MAX base their model certainty on the amount of collected data, while Bayesian approaches assume a prior over the transition dynamics. We propose extensions to such approaches which drive exploration solely based on empirical estimates of the learner's accuracy and Learning Progress. We provide a "sanity check" theoretical analysis, discussing the behavior of our extensions in the standard stationary finite state-action case. We then provide experimental studies demonstrating the robustness of these exploration measures in cases of non-stationary environments or where original approaches are misled by wrong domain assumptions.
Yu Ogura - One of the best experts on this subject based on the ideXlab platform.
-
development of a suture ligature training system designed to provide quantitative information of the Learning Progress of trainees
International Conference on Robotics and Automation, 2007Co-Authors: N Oshima, M Aizudding, R Midorikawa, Jorge Solis, Yu Ogura, Atsuo TakanishiAbstract:Surgeons, during a medical intervention, perform different kinds of manual tasks with dexterity and precision. In order to assure the success of the intervention, medical students are trained several years using training models so that they can perform such tasks with accuracy to avoid any possible risk to patients. However, current training models are merely designed to imitate the surgical procedure without providing any further information about the how well the task was done. For that reason; in this paper, the development of a suture/ligature training system is proposed to imitate surgical procedures as well as provide quantitative information of the Learning Progress of trainees. As a first approach, a training system was designed to simulate the suture and ligature tasks. The proposed training system includes a skin dummy with an array of embedded photo interrupters to detect the movement of the skin dummy; without requiring any modification on the surgical instrument. In this paper, different task parameters were proposed to understand trainees' Learning Progress and different experiments were carried out to identify the evaluation parameters that may provide useful information about how well the task was performed. As a result from the experiments, the evaluation parameters for the suture and ligature were determined. Finally, an evaluation function was proposed and further experiments were proposed to verify its effectiveness. From the results of the experiments, we could effectively distinguish quantitatively the differences of skill levels between surgeons and unskilled persons as well as identifying the Learning Progress of trainees by plotting the Learning curve
-
ICRA - Development of a Suture/Ligature Training System designed to provide quantitative information of the Learning Progress of trainees
Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007Co-Authors: N Oshima, M Aizudding, R Midorikawa, Jorge Solis, Yu Ogura, Atsuo TakanishiAbstract:Surgeons, during a medical intervention, perform different kinds of manual tasks with dexterity and precision. In order to assure the success of the intervention, medical students are trained several years using training models so that they can perform such tasks with accuracy to avoid any possible risk to patients. However, current training models are merely designed to imitate the surgical procedure without providing any further information about the how well the task was done. For that reason; in this paper, the development of a suture/ligature training system is proposed to imitate surgical procedures as well as provide quantitative information of the Learning Progress of trainees. As a first approach, a training system was designed to simulate the suture and ligature tasks. The proposed training system includes a skin dummy with an array of embedded photo interrupters to detect the movement of the skin dummy; without requiring any modification on the surgical instrument. In this paper, different task parameters were proposed to understand trainees' Learning Progress and different experiments were carried out to identify the evaluation parameters that may provide useful information about how well the task was performed. As a result from the experiments, the evaluation parameters for the suture and ligature were determined. Finally, an evaluation function was proposed and further experiments were proposed to verify its effectiveness. From the results of the experiments, we could effectively distinguish quantitatively the differences of skill levels between surgeons and unskilled persons as well as identifying the Learning Progress of trainees by plotting the Learning curve
