The Experts below are selected from a list of 111 Experts worldwide ranked by ideXlab platform
Donald C. Wunsch - One of the best experts on this subject based on the ideXlab platform.
-
Safe Intermittent Reinforcement Learning for Nonlinear Systems
2019 IEEE 58th Conference on Decision and Control (CDC), 2019Co-Authors: Yongliang Yang, Kyriakos G. Vamvoudakis, Hamidreza Modares, Wei He, Donald C. WunschAbstract:In this paper, an online Intermittent actor-critic Reinforcement learning method is used to stabilize nonlinear systems optimally while also guaranteeing safety. A barrier function-based transformation is introduced to ensure that the system does not violate the user-defined safety constraints. It is shown that the safety constraints of the original system can be guaranteed by assuring the stability of the equilibrium point of an appropriately transformed system. Then, an online Intermittent actor-critic learning framework is developed to learn the optimal safe Intermittent controller. Also, Zeno behavior is guaranteed to be excluded. Finally, numerical examples are conducted to verify the efficacy of the learning algorithm.
-
Safe Intermittent Reinforcement Learning With Static and Dynamic Event Generators
IEEE Transactions on Neural Networks and Learning Systems, 1Co-Authors: Yongliang Yang, Kyriakos G. Vamvoudakis, Hamidreza Modares, Donald C. WunschAbstract:In this article, we present an Intermittent framework for safe Reinforcement learning (RL) algorithms. First, we develop a barrier function-based system transformation to impose state constraints while converting the original problem to an unconstrained optimization problem. Second, based on optimal derived policies, two types of Intermittent feedback RL algorithms are presented, namely, a static and a dynamic one. We finally leverage an actor/critic structure to solve the problem online while guaranteeing optimality, stability, and safety. Simulation results show the efficacy of the proposed approach.
George H. Noell - One of the best experts on this subject based on the ideXlab platform.
-
Programming for Maintenance: An Investigation of Delayed Intermittent Reinforcement and Common Stimuli to Create Indiscriminable Contingencies
Journal of Behavioral Education, 2002Co-Authors: Jennifer T. Freeland, George H. NoellAbstract:Generalization across time or maintenance of behavior change is a fundamental concern for behavior analysts and educators that remains insufficiently understood. This study examined the maintenance of mathematics responding during and following delayed Intermittent Reinforcement when common stimuli were programmed across the treatment and maintenance phases. Two third-grade girls who were referred by their classroom teacher due to concerns in the area of mathematics participated. Students were exposed to baseline, contingent Reinforcement, delayed Intermittent Reinforcement, and a maintenance condition. The maintenance condition followed exposure to delayed Intermittent Reinforcement and included common stimuli from the Reinforcement condition, but did not include a contingency for correct responding. Both students exhibited substantial prolonged maintenance during this condition. Implications of these results for future research examining maintenance and applied programming for maintenance are discussed.
-
maintaining accurate math responses in elementary school students the effects of delayed Intermittent Reinforcement and programming common stimuli
Journal of Applied Behavior Analysis, 1999Co-Authors: Jennifer T. Freeland, George H. NoellAbstract:This study examined the effect of delayed Reinforcement on digits completed by elementary school children and the effect of programming stimuli common to Reinforcement conditions on the maintenance of their performance. Participants exhibited similar levels of responding during Intermittent and continuous Reinforcement. Responding continued for a number of sessions at similar levels during a maintenance phase that included stimuli present during delayed Reinforcement.
Yongliang Yang - One of the best experts on this subject based on the ideXlab platform.
-
Safe Intermittent Reinforcement Learning for Nonlinear Systems
2019 IEEE 58th Conference on Decision and Control (CDC), 2019Co-Authors: Yongliang Yang, Kyriakos G. Vamvoudakis, Hamidreza Modares, Wei He, Donald C. WunschAbstract:In this paper, an online Intermittent actor-critic Reinforcement learning method is used to stabilize nonlinear systems optimally while also guaranteeing safety. A barrier function-based transformation is introduced to ensure that the system does not violate the user-defined safety constraints. It is shown that the safety constraints of the original system can be guaranteed by assuring the stability of the equilibrium point of an appropriately transformed system. Then, an online Intermittent actor-critic learning framework is developed to learn the optimal safe Intermittent controller. Also, Zeno behavior is guaranteed to be excluded. Finally, numerical examples are conducted to verify the efficacy of the learning algorithm.
-
Safe Intermittent Reinforcement Learning With Static and Dynamic Event Generators
IEEE Transactions on Neural Networks and Learning Systems, 1Co-Authors: Yongliang Yang, Kyriakos G. Vamvoudakis, Hamidreza Modares, Donald C. WunschAbstract:In this article, we present an Intermittent framework for safe Reinforcement learning (RL) algorithms. First, we develop a barrier function-based system transformation to impose state constraints while converting the original problem to an unconstrained optimization problem. Second, based on optimal derived policies, two types of Intermittent feedback RL algorithms are presented, namely, a static and a dynamic one. We finally leverage an actor/critic structure to solve the problem online while guaranteeing optimality, stability, and safety. Simulation results show the efficacy of the proposed approach.
Jennifer T. Freeland - One of the best experts on this subject based on the ideXlab platform.
-
Programming for Maintenance: An Investigation of Delayed Intermittent Reinforcement and Common Stimuli to Create Indiscriminable Contingencies
Journal of Behavioral Education, 2002Co-Authors: Jennifer T. Freeland, George H. NoellAbstract:Generalization across time or maintenance of behavior change is a fundamental concern for behavior analysts and educators that remains insufficiently understood. This study examined the maintenance of mathematics responding during and following delayed Intermittent Reinforcement when common stimuli were programmed across the treatment and maintenance phases. Two third-grade girls who were referred by their classroom teacher due to concerns in the area of mathematics participated. Students were exposed to baseline, contingent Reinforcement, delayed Intermittent Reinforcement, and a maintenance condition. The maintenance condition followed exposure to delayed Intermittent Reinforcement and included common stimuli from the Reinforcement condition, but did not include a contingency for correct responding. Both students exhibited substantial prolonged maintenance during this condition. Implications of these results for future research examining maintenance and applied programming for maintenance are discussed.
-
maintaining accurate math responses in elementary school students the effects of delayed Intermittent Reinforcement and programming common stimuli
Journal of Applied Behavior Analysis, 1999Co-Authors: Jennifer T. Freeland, George H. NoellAbstract:This study examined the effect of delayed Reinforcement on digits completed by elementary school children and the effect of programming stimuli common to Reinforcement conditions on the maintenance of their performance. Participants exhibited similar levels of responding during Intermittent and continuous Reinforcement. Responding continued for a number of sessions at similar levels during a maintenance phase that included stimuli present during delayed Reinforcement.
Hamidreza Modares - One of the best experts on this subject based on the ideXlab platform.
-
Safe Intermittent Reinforcement Learning for Nonlinear Systems
2019 IEEE 58th Conference on Decision and Control (CDC), 2019Co-Authors: Yongliang Yang, Kyriakos G. Vamvoudakis, Hamidreza Modares, Wei He, Donald C. WunschAbstract:In this paper, an online Intermittent actor-critic Reinforcement learning method is used to stabilize nonlinear systems optimally while also guaranteeing safety. A barrier function-based transformation is introduced to ensure that the system does not violate the user-defined safety constraints. It is shown that the safety constraints of the original system can be guaranteed by assuring the stability of the equilibrium point of an appropriately transformed system. Then, an online Intermittent actor-critic learning framework is developed to learn the optimal safe Intermittent controller. Also, Zeno behavior is guaranteed to be excluded. Finally, numerical examples are conducted to verify the efficacy of the learning algorithm.
-
Safe Intermittent Reinforcement Learning With Static and Dynamic Event Generators
IEEE Transactions on Neural Networks and Learning Systems, 1Co-Authors: Yongliang Yang, Kyriakos G. Vamvoudakis, Hamidreza Modares, Donald C. WunschAbstract:In this article, we present an Intermittent framework for safe Reinforcement learning (RL) algorithms. First, we develop a barrier function-based system transformation to impose state constraints while converting the original problem to an unconstrained optimization problem. Second, based on optimal derived policies, two types of Intermittent feedback RL algorithms are presented, namely, a static and a dynamic one. We finally leverage an actor/critic structure to solve the problem online while guaranteeing optimality, stability, and safety. Simulation results show the efficacy of the proposed approach.