Convergence of simultaneous perturbation stochastic approximation for nondifferentiable optimization

We consider simultaneous perturbation stochastic approximation for function minimization. The standard assumption for convergence is that the function be three times differentiable, although weaker assumptions have been used for special cases. However, all work that we are aware of at least requires differentiability. We relax the differentiability requirement and prove convergence using convex analysis.     

Nicotine withdrawal in chippers and regular smokers: Subjective and cognitive effects.

From previous studies, chippers (very light, long-time cigarette smokers) seem not to be nicotine dependent, despite decades of smoking. The effect of tobacco deprivation on chippers' withdrawal reactions was examined. Matched groups of 26 chippers and 25 regular smokers were studied while abstaining or smoking for 2-day blocks, with assessments administered 5 times daily by palm-top computers. As hypothesized, chippers showed no changes as a result of nicotine deprivation. In contrast, regular smokers demonstrated distinct changes in craving, mood, arousal, and sleep disturbance. The computers also tested participants' cognitive performance. Unlike chippers, regular smokers' performance on complex tasks was slower under deprivation; the effect could not be explained by changes in motor performance or simple reaction time. Results suggest performance may have been improved by nicotine rather than worsened by withdrawal. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Bodies, antibodies, and neighborhood-density effects in masked form priming.

Facilitatory priming effects due to similarity of orthographic form are obtained for high-N target words provided that they have low-frequency bodies and the body is shared between the prime and target (e.g., perd-HERD). Conversely, it is shown that low-N target words show priming regardless of the frequency of the body, provided that the prime and target do not share the same body (e.g., drice-DRIVE). If the body is shared, then priming occurs only for targets with low-frequency bodies. These results suggest that neighborhood density should be defined in terms of both individual letter units and subsyllabic units and that both types of density jointly determine priming. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Peer-tutoring: Applied behavior analysis working in the classroom.

Schools often use peer-tutoring as a corrective program for students who are experiencing difficulties academically. Given appropriate behavioral structure, peer-tutoring can become a powerful classwide instructional intervention. An historical perspective on the use of students as teachers for their peers is given in this selective literature review. Also, information on the efficacy of considering behavioral function and structure and data addressing ecobehavioral variables to establish a productive learning environment are provided. The authors emphasize infusing peer-tutoring programs with applied behavior analysis techniques to increase the academic success of large numbers of students cost effectively. Peer-tutoring programs are considered to be a way for applied behavior analysts to gain greater access to American school learning environments. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Synthesizing Energy-Efficient Embedded Systems with LOPOCOS

In this paper, we introduce the LOPOCOS (Low Power Co-synthesis) system, a prototype CAD tool for system level co-design. LOPOCOS targets the design of energy-efficient embedded systems implemented as heterogeneous distributed architectures. In particular, it is designed to solve the specific problems involved in architectures that include dynamic voltage scalable (DVS) processors. The aim of this paper is to demonstrate how LOPOCOS can support the system designer in identifying energy-efficient hardware/software implementations for the desired embedded systems. Hence, highlighting the necessary optimization steps during design space exploration for DVS enable architectures. The optimization steps carried out in LOPOCOS involve component allocation and task/communication mapping as well as scheduling and dynamic voltage scaling. LOPOCOS has the following key features, which contribute to this energy efficiency. During the voltage scaling valuable power profile information of task execution is taken into account, hence, the accuracy of the energy estimation is improved. A combined optimization for scheduling and communication mapping based on genetic algorithm, optimizes simultaneously execution order and communication mapping towards the utilization of the DVS processors and timing behaviour. Furthermore, a separation of task and communication mapping allows a more effective implementation of both task and communication mapping optimizationsteps. Extensive experiments are conducted to demonstrate the efficiency of LOPOCOS. We report up to 38% higher energy reductions compared to previous co-synthesis techniques for DVS systems. The investigations include a real-life example of an optical flow detection algorithm.     

THE FEDERAL ROLE IN ENERGY R&D

This paper documents and analyzes recent trends in Federal energy research and development programs. In light of the marketplace philosophy of the present Administration, the possible benefits of a strong Federal research program versus the potential advantages of relying on private sector research programs to develop new technology alternatives are analyzed in some detail. The impacts of the Reagan budget cuts and other initiatives on several major energy programs, including nuclear power, photovoltaics, and conservation, are also analyzed. This analysis reveals some of the difficulties inherent in attempting to project the consequences of particular policy decisions in a complex and changing environment, and identifies some of the problems of trying to evaluate the success of particular programs or actions on a limited time scale.     

Longitudinal investigation of the abstinence violation effect in binge eaters.

This study, guided in part by G. A. Marlatt and J. R. Gordon's (1985) abstinence violation effect (AVE) model, examined whether variability in cognitive and emotional reactions to binges accounts for recurrence of binge eating. Attributional, cognitive, and affective reactions to 2 successive binges, as well as the latencies between each binge and a subsequent binge, were evaluated through a series of structured phone interviews with 50 nonpurging normal-weight female bingers. Reported mood after binging was unrelated to binge latency. However, when Ss made more intense internal, global, and uncontrollable causal attributions for a binge, a subsequent binge followed significantly sooner. Within-subject variations in AVEs across binges prospectively predicted within-subject variations in the speed with which another binge followed. Analyses suggested that cognitive states (e.g., AVE and guilt) evoked by particular events were better predictors of how quickly binging repeats than were stable differences in attributional style. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The relationship between the extent of reciprocity with social supporters and measures of depressive symptomatology, impairment, disability, and handicap in persons with spinal cord injury.

140 adults (aged 19–77 yrs) were randomly selected from a sampling frame of 661 persons with spinal cord injury. They completed a questionnaire assessing amount of social support, degree of reciprocity with supporters, and satisfaction with relationships with supporters; the Center for Epidemiologic Studies Depression Scale; a self-report version of the Functional Independence Measure; and the Craig Handicap Assessment and Reporting Technique. A physician assessed the degree of paralytic impairment by means of the ASIA Total Motor Index Score. Relationships in which the supporter gave more than the participant were associated with more satisfaction with the relationship and more social support. The number of reciprocal relationships was positively associated with more desirable outcomes in the handicap dimensions of mobility, occupation, social integration, and economic self-sufficiency. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Optimal block-type-decodable encoders for constrained systems

A constrained system is presented by a finite-state labeled graph. For such systems, we focus on block-type-decodable encoders, comprising three classes known as block, block-decodable, and deterministic encoders. Franaszek (1968) gives a sufficient condition which guarantees the equality of the optimal rates of block-decodable and deterministic encoders for the same block length. We introduce another sufficient condition, called the straight-line condition, which yields the same result. Run-length limited RLL(d,k) and maximum transition run MTR(j,k) constraints are shown to satisfy both conditions. In general, block-type-decodable encoders are constructed by choosing a subset of states of the graph to be used as encoder states. Such a subset is known as a set of principal states. For each type of encoder and each block length, a natural problem is to find a set of principal states which maximizes the code rate. We show how to compute the asymptotically optimal sets of principal states for deterministic encoders and how they are related to the case of large but finite block lengths. We give optimal sets of principal states for MTR(j,k)-block-type-decodable encoders for all codeword lengths. Finally we compare the code rate of nonreturn to zero inverted (NRZI) encoders to that of corresponding nonreturn to zero (NRZ) and signed NRZI encoders.