The Experts below are selected from a list of 5805 Experts worldwide ranked by ideXlab platform
Sebastian Otte - One of the best experts on this subject based on the ideXlab platform.
-
learning planning and control in a monolithic neural event inference architecture
Neural Networks, 2019Co-Authors: Martin V Butz, David K Bilkey, Dania Humaidan, Alistair Knott, Sebastian OtteAbstract:We introduce REPRISE, a REtrospective and PRospective Inference SchEme, which learns temporal event-predictive models of dynamical systems. REPRISE infers the unobservable contextual event state and accompanying temporal predictive models that best explain the recently encountered sensorimotor experiences retrospectively. Meanwhile, it optimizes upcoming motor activities prospectively in a goal-directed manner. Here, REPRISE is implemented by a recurrent neural network (RNN), which learns temporal forward models of the sensorimotor contingencies generated by different simulated dynamic vehicles. The RNN is augmented with contextual neurons, which enable the encoding of distinct, but related, sensorimotor dynamics as compact event codes. We show that REPRISE concurrently learns to separate and approximate the encountered sensorimotor dynamics: it analyzes sensorimotor error signals adapting both internal contextual neural activities and connection weight values. Moreover, we show that REPRISE can exploit the learned model to induce goal-directed, model-predictive control, that is, approximate active inference: Given a goal state, the system imagines a motor Command Sequence optimizing it with the prospective objective to minimize the distance to the goal. The RNN activities thus continuously imagine the upcoming future and reflect on the recent past, optimizing the predictive model, the hidden neural state activities, and the upcoming motor activities. As a result, event-predictive neural encodings develop, which allow the invocation of highly effective and adaptive goal-directed sensorimotor control.
Sarah Pulley Basile - One of the best experts on this subject based on the ideXlab platform.
-
Using a High Probability Command Sequence to Increase Classroom Compliance: The Role of Behavioral Momentum
Journal of Behavioral Education, 2008Co-Authors: Phillip J. Belfiore, Sarah Pulley BasileAbstract:One of the most problematic behaviors in children with developmental disabilities is noncompliance. Although behavioral research has provided strategies to impact noncompliance, oftentimes the methodologies are consequent techniques, which may not be conducive to implementation by the classroom teacher. In this teacher-designed and implemented study, a Sequence of high-probability instructional Commands preceded the targeted low-probability Command, in an attempt to increase compliance to the low-probability Command. Results, discussed within the body of behavioral momentum research, showed an increase in compliance to low-probability classroom Commands for a seven year-old student with moderate mental retardation and Down Syndrome. Results are discussed as (a) an effective, antecedent approach to classroom compliance and (b) re-connecting the gap between applied behavioral research and experimentally controlled classroom practice.
Amber J. Zank - One of the best experts on this subject based on the ideXlab platform.
-
Increasing Classroom Compliance: Using a High-Probability Command Sequence with Noncompliant Students
Journal of Behavioral Education, 2012Co-Authors: Michael I. Axelrod, Amber J. ZankAbstract:Noncompliance is one of the most problematic behaviors within the school setting. One strategy to increase compliance of noncompliant students is a high-probability Command Sequence (HPCS; i.e., a set of simple Commands in which an individual is likely to comply immediately prior to the delivery of a Command that has a lower probability of compliance). Although research has shown this technique to be effective at increasing compliance across various settings and behaviors, most studies have been limited to participants with moderate to severe developmental disabilities. The current study targeted 2 noncompliant elementary-age students within the general education setting. Two teachers were taught to integrate HPCS into ongoing classroom reading instruction and independent seatwork. For both participants, higher percentages of compliance with low-probability Commands were displayed during intervention and maintenance phases compared to baseline levels. Results suggest that using an antecedent intervention based on HPCS holds promise for school personnel working with noncompliant students within the general education setting.
Martin V Butz - One of the best experts on this subject based on the ideXlab platform.
-
learning planning and control in a monolithic neural event inference architecture
Neural Networks, 2019Co-Authors: Martin V Butz, David K Bilkey, Dania Humaidan, Alistair Knott, Sebastian OtteAbstract:We introduce REPRISE, a REtrospective and PRospective Inference SchEme, which learns temporal event-predictive models of dynamical systems. REPRISE infers the unobservable contextual event state and accompanying temporal predictive models that best explain the recently encountered sensorimotor experiences retrospectively. Meanwhile, it optimizes upcoming motor activities prospectively in a goal-directed manner. Here, REPRISE is implemented by a recurrent neural network (RNN), which learns temporal forward models of the sensorimotor contingencies generated by different simulated dynamic vehicles. The RNN is augmented with contextual neurons, which enable the encoding of distinct, but related, sensorimotor dynamics as compact event codes. We show that REPRISE concurrently learns to separate and approximate the encountered sensorimotor dynamics: it analyzes sensorimotor error signals adapting both internal contextual neural activities and connection weight values. Moreover, we show that REPRISE can exploit the learned model to induce goal-directed, model-predictive control, that is, approximate active inference: Given a goal state, the system imagines a motor Command Sequence optimizing it with the prospective objective to minimize the distance to the goal. The RNN activities thus continuously imagine the upcoming future and reflect on the recent past, optimizing the predictive model, the hidden neural state activities, and the upcoming motor activities. As a result, event-predictive neural encodings develop, which allow the invocation of highly effective and adaptive goal-directed sensorimotor control.
Phillip J. Belfiore - One of the best experts on this subject based on the ideXlab platform.
-
Using a High Probability Command Sequence to Increase Classroom Compliance: The Role of Behavioral Momentum
Journal of Behavioral Education, 2008Co-Authors: Phillip J. Belfiore, Sarah Pulley BasileAbstract:One of the most problematic behaviors in children with developmental disabilities is noncompliance. Although behavioral research has provided strategies to impact noncompliance, oftentimes the methodologies are consequent techniques, which may not be conducive to implementation by the classroom teacher. In this teacher-designed and implemented study, a Sequence of high-probability instructional Commands preceded the targeted low-probability Command, in an attempt to increase compliance to the low-probability Command. Results, discussed within the body of behavioral momentum research, showed an increase in compliance to low-probability classroom Commands for a seven year-old student with moderate mental retardation and Down Syndrome. Results are discussed as (a) an effective, antecedent approach to classroom compliance and (b) re-connecting the gap between applied behavioral research and experimentally controlled classroom practice.