Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Probability Learning”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,459 records · Page 81Linked to original sources
PubMed2000

You can go home again: evidence from longitudinal data.

In this paper we analyze the economic and demographic factors that influence return migration, focusing on generation 1.5 immigrants. Using longitudinal data from the 1979 youth cohort of the National Longitudinal Surveys (NLSY79), we track residential histories of young immigrants to the United States and analyze the covariates associated with return migration to their home country. Overall, return migration appears to respond to economic incentives, as well as to cultural and linguistic ties to the United States and the home country. We find no role for welfare magnets in the decision to return, but we learn that welfare participation leads to lower probability of return migration. Finally, we see no evidence of a skill bias in return migration, where skill is measured by performance on the Armed Forces Qualifying Test.

Age Factors↗
PubMed2004

[New aspects in the management of renal osteodystrophy].

Our knowledge of the mineral metabolism disturbances and skeletal disorders in patients with chronic renal insufficiency has advanced significantly in the last years. Probably the most important what we have learned is, that apart of bone disease, hyperparathyroidism and its treatment can also lead to severe extra-skeletal complications, contributing to the progression of cardiovascular disease in this population. This fact has fundamentally changed our approach to the treatment, and - as a result - the new clinical practice guidelines on the management of mineral metabolism and bone disease in chronic renal failure have been developed. Mainstays of the new approach are lower recommended serum calcium and calcium-phosphate product. However, these targets are very difficult to achieve in clinical practice. During the last decade, some additional therapeutic agents have been welcomed. There are two effective calcium-free, aluminium-free phosphate binders, sevelamer hydrochloride, and lanthanum carbonate, four vitamin D analogues of newer generation, and--last but not least--modulators of calcium-sensing receptor, calcimimetics. Future studies will be needed to determine how these recent developments will be helpful. The optimal therapy for all end-stage renal failure complications is definitely renal transplantation.

Calcinosis↗
PubMed2006

Multiple roles of experience in decoding the neural representation of sensory stimuli.

Experience and perception are deeply intertwined. Experience, particularly early in life, shapes how sensory information is represented in the brain. Experience also establishes associations and can affect how sensory information guides behaviour. Central to these kinds of perceptual abilities are neural mechanisms that interpret, or decode, the brain's sensory representation, but little is known about how these decoding mechanisms depend on experience. Here I discuss several critical roles that experience might play in shaping these mechanisms. First, experience is likely to drive changes in neural connectivity to select the spatially and temporally distributed sensory signals that provide relevant information about a stimulus. Second, even the most relevant sensory signals provide incomplete information about the presence of a stimulus; also necessary is knowledge of the a priori probability of the stimulus, which must be learned from experience. Third, decoding noisy information is necessarily imperfect and therefore involves trade-offs like speed versus accuracy and false alarms versus misses. Experience is likely to provide ongoing feedback about the value of these trade-offs so that they might be adjusted appropriately. Each of these mechanisms appear to be capable of causing dramatic changes in sensitivity, response bias, response times and other manifestations of perceptual ability.

Animals↗
PubMed1991

Violence in a pediatric emergency department: lessons learned from one experience.

A multitude of forces influence the probability of violence in an emergency department, but none impact as directly as the skilled response of the emergency nurse. It is important not to overlook the availability of security and law enforcement, but it may be equally important for nurses to take appropriate steps toward preventing or solving such problems with a minimum of physical and emotional pain for all concerned. With time and practice, incidents of violence can be replaced with opportunities for growth and healing.

Child, Preschool↗
PubMed2000

Effects of high-probability requests on the latency to initiate academic tasks.

The purpose of this study was to evaluate the effectiveness of a high-probability request sequence on the latency to and duration of compliance to a request for completion of an independent math assignment. The participant was an elementary-school student with learning disabilities who exhibited noncompliance during math instruction. The results showed that high-probability requests were effective in reducing the latency to compliance but only minimally affected duration of engagement.

Adolescent↗
PubMedDate not supplied

Bayesian applications of belief networks and multilayer perceptrons for ovarian tumor classification with rejection.

Incorporating prior knowledge into black-box classifiers is still much of an open problem. We propose a hybrid Bayesian methodology that consists in encoding prior knowledge in the form of a (Bayesian) belief network and then using this knowledge to estimate an informative prior for a black-box model (e.g. a multilayer perceptron). Two technical approaches are proposed for the transformation of the belief network into an informative prior. The first one consists in generating samples according to the most probable parameterization of the Bayesian belief network and using them as virtual data together with the real data in the Bayesian learning of a multilayer perceptron. The second approach consists in transforming probability distributions over belief network parameters into distributions over multilayer perceptron parameters. The essential attribute of the hybrid methodology is that it combines prior knowledge and statistical data efficiently when prior knowledge is available and the sample is of small or medium size. Additionally, we describe how the Bayesian approach can provide uncertainty information about the predictions (e.g. for classification with rejection). We demonstrate these techniques on the medical task of predicting the malignancy of ovarian masses and summarize the practical advantages of the Bayesian approach. We compare the learning curves for the hybrid methodology with those of several belief networks and multilayer perceptrons. Furthermore, we report the performance of Bayesian belief networks when they are allowed to exclude hard cases based on various measures of prediction uncertainty.

Bayes Theorem↗
PubMed1997

Symmetry breaking and training from incomplete data with Radial Basis Boltzmann Machines.

A Radial Basis Boltzmann Machine (RBBM) is a specialized Boltzmann Machine architecture that combines feed-forward mapping with probability estimation in the input space, and for which very efficient learning rules exist. The hidden representation of the network displays symmetry breaking as a function of the noise in the dynamics. Thus, generalization can be studied as a function of the noise in the neuron dynamics instead of as a function of the number of hidden units. We show that the RBBM can be seen as an elegant alternative of k-nearest neighbor, leading to comparable performance without the need to store all data. We show that the RBBM has good classification performance compared to the MLP. The main advantage of the RBBM is that simultaneously with the input-output mapping, a model of the input space is obtained which can be used for learning with missing values. We derive learning rules for the case of incomplete data, and show that they perform better on incomplete data than the traditional learning rules on a 'repaired' data set.

Computer Simulation↗
PubMed1967

Effects of reinforcement scheduling on simultaneous discrimination performance.

Pigeons were trained on a discrete-trials, simultaneous discrimination procedure, with confusable stimuli such that asymptotic performance was about 85% correct. Trials were terminated if no response occurred within 2 sec of stimulus onset, so that probability of responding was free to vary. The schedule of reinforcement for correct responses was varied, with the following results: (1) there was no relation between frequency of reinforcement and accuracy of responding. (2) In extinction, the probability of responding fell to low levels, but accuracy remained roughly constant. (3) When reinforcement was available after a fixed number of trials or after a fixed number of correct responses, the probability of responding increased with successive trials after reinforcement, but accuracy was generally constant. (4) When every fifth correct response was reinforced, accuracy decreased immediately after reinforcement if the birds were required to respond on every trial.

Animals↗
PubMed1985

Reversal learning in senescent rats.

The ability of old (24 months) and young (3 months) male rats to reverse a previously acquired discrimination was compared in 5 experiments. The old rats did not need more trials to learn a position habit in a T-maze to obtain water reward, but required more trials to reverse the position habit. The old rats showed a similar deficit in a second, but not in subsequent reversals of the position habit. In a second experiment, old rats were slower in learning to operate one of two levers in an operant chamber to obtain food reward on a CRF schedule, but by the session prior to reaching criterion for acquisition they showed response rates similar to the young animals. When the rats were required to operate the alternative lever to obtain reward, the young rats emitted 70% of their responses during the first reversal session on the newly-correct lever, but the old rats only 35%. Nevertheless, the groups were similar in the number of sessions required to reach a criterion of 95% of responses on the correct lever. In 3 subsequent reversals, old and young rats did not differ nor were there differences in the number of responses in 4 extinction sessions in the rats which had received reversal training. In experiment 3 with old and young rats which had received only acquisition training, old rats emitted fewer responses than young animals during extinction. From these experiments it was hypothesized that the apparent difficulty of old rats in learning a reversal task was due to the low probability of their emitting spontaneously a novel or previously unrewarded response, and not to a difficulty in forming a new association. This hypothesis was tested in two further experiments in which rats were required to learn a brightness discrimination in a T-maze. Old and young rats which had learned and reversed position habits in the T-maze in experiment 1, did not differ in either acquisition or reversal of the brightness discrimination, suggesting that old rats do not differ from young animals in reversal tasks when the motor response requirements for the task are already within the animals' behavioural repertoire. Consistent with this hypothesis, naive old rats were slower than young rats in acquiring a similar brightness discrimination but did not differ in the reversal task.

Aging↗
PubMedDate not supplied

Everything I needed to know about medical management I learned in acting school.

Some people are doctors. Some people play them on TV. Brian Meltzer could probably do both. A physician executive at Memorial Sloan-Kettering Cancer Center, Meltzer relies on skills he learned in acting school to help manage business decisions. In the first of several essays for The Physician Executive, Meltzer explains how acting can help you become a better leader.

Anecdotes as Topic↗
PubMed1999

Ibotenate lesions of the hippocampus impair spatial learning but not contextual fear conditioning in mice.

Recently, gene targeting and other mouse transgenic techniques have been used to study the cellular mechanisms underlying learning and memory mechanisms in the hippocampus. A key assumption of many of these studies is that lesions of the hippocampus have a similar impact on learning and memory in mice and in rats. Here, we used axon-sparing ibotenate lesions to determine whether damage to the hippocampus disrupts spatial learning and contextual conditioning in mice, as it is known to do in rats. Our results demonstrated that hippocampal lesions impair performance in the hidden-platform version of the water maze under a variety of experimental conditions. Neither keeping the start site constant, nor prior training with the visible-platform task fully rescued the spatial learning deficits of the lesioned mice. As previously shown in rats, the lesions left the performance of the mice intact in the visible-platform version of the water maze, indicating that they do not affect all types of learning, and that disruptions of sensory processing or motivation probably did not account for their deficits in the hidden-platform task. In contrast, the very same lesions did not affect either cued or contextual fear conditioning. These results confirm the involvement of the hippocampus in spatial learning in mice, and they also demonstrate that hippocampal-lesioned mice can show contextual fear conditioning. Thus, the behavioral findings presented here are crucial for the interpretation of transgenic experiments with the widely used water maze and fear-conditioning paradigms.

Animals↗
PubMed2001

Algebraic geometrical methods for hierarchical learning machines.

Hierarchical learning machines such as layered perceptrons, radial basis functions, Gaussian mixtures are non-identifiable learning machines, whose Fisher information matrices are not positive definite. This fact shows that conventional statistical asymptotic theory cannot be applied to neural network learning theory, for example either the Bayesian a posteriori probability distribution does not converge to the Gaussian distribution, or the generalization error is not in proportion to the number of parameters. The purpose of this paper is to overcome this problem and to clarify the relation between the learning curve of a hierarchical learning machine and the algebraic geometrical structure of the parameter space. We establish an algorithm to calculate the Bayesian stochastic complexity based on blowing-up technology in algebraic geometry and prove that the Bayesian generalization error of a hierarchical learning machine is smaller than that of a regular statistical model, even if the true distribution is not contained in the parametric model.

Algorithms↗
PubMed2005

Directed forgetting in incidental learning and recognition testing: support for a two-factor account.

Instructing people to forget a list of items often leads to better recall of subsequently studied lists (known as the benefits of directed forgetting). The authors have proposed that changes in study strategy are a central cause of the benefits (L. Sahakyan & P. F. Delaney, 2003). The authors address 2 results from the literature that are inconsistent with their strategy-based explanation: (a) the presence of benefits under incidental learning conditions and (b) the absence of benefits in recognition testing. Experiment 1 showed that incidental learning attenuated the benefits compared with intentional learning, as expected if a change of study strategy causes the benefits. Experiment 2 demonstrated benefits using recognition testing, albeit only when longer lists were used. Memory for source in directed forgetting was also explored using multinomial modeling. Results are discussed in terms of a 2-factor account of directed forgetting.

Attention↗
PubMed1997

Probabilistic incremental program evolution

Probabilistic incremental program evolution (PIPE) is a novel technique for automatic program synthesis. We combine probability vector coding of program instructions, population-based incremental learning, and tree-coded programs like those used in some variants of genetic programming (GP). PIPE iteratively generates successive populations of functional programs according to an adaptive probability distribution over all possible programs. Each iteration, it uses the best program to refine the distribution. Thus, it stochastically generates better and better programs. Since distribution refinements depend only on the best program of the current population, PIPE can evaluate program populations efficiently when the goal is to discover a program with minimal runtime. We compare PIPE to GP on a function regression problem and the 6-bit parity problem. We also use PIPE to solve tasks in partially observable mazes, where the best programs have minimal runtime.

Journal Article↗
PubMed1986

Automatic activation of episodic information in a semantic memory task.

Four experiments are presented in which priming between newly learned associates and priming between well-known associates were examined in lexical decision. All four experiments found priming between newly learned associates, including conditions in which the Stimulus Onset Asynchrony (SOA) between prime and target was short (150 ms) and in which the probability was low (1/12) that the prime and target of a pair would be associated to each other. It was concluded, contrary to suggestions by Carroll and Kirsner (1982) and Tulving (1983), that newly learned associates can prime each other, and that they can do so at short SOAs.

Association Learning↗
PubMed1993

Can restenosis after coronary angioplasty be predicted from clinical variables?

OBJECTIVES: The purpose of this study was to determine whether variables shown to correlate with restenosis in one group (learning group) could be shown to predict recurrent stenosis in a second group (validation group). BACKGROUND: Restenosis remains a critical limitation after percutaneous transluminal coronary angioplasty. Although several clinical variables have been shown to correlate with restenosis, there are few data concerning attempts to predict recurrent stenosis. METHODS: The source of data was the clinical data base at Emory University. Patients who had had previous coronary surgery and patients who underwent coronary angioplasty in the setting of acute myocardial infarction were excluded. A total of 4,006 patients with angiographic restudy after successful angioplasty were identified. They were classified into a learning group of 2,500 patients and a validation group of 1,506 patients. The correlates of restenosis in the learning group were determined by stepwise logistic regression, and a model was developed to predict the probability of restenosis and was tested in the validation group. By using various cut points for the predicted probability of restenosis, a receiver operating characteristic curve was created. Goodness of fit of the model was evaluated by comparing average predicted probabilities with average observed probabilities within subgroups on the basis of risk level determined by linear regression analysis. RESULTS: In the learning group 1,145 patients had restenosis and 1,355 did not. Correlates of restenosis were severe angina, severe diameter stenosis before angioplasty, left anterior descending coronary artery dilation, diabetes, greater diameter stenosis after angioplasty, hypertension, absence of an intimal tear, eccentric morphology and older patient age. The model derived from the learning group was used to predict restenosis in the validation group. By varying the cut point for the predicted probability of restenosis above which restenosis is diagnosed and below which it is not, a receiver operating characteristic curve was created. The curve was close to the line of identity, reflecting a poor predictive ability. However, the model was shown to fit well with the predicted probability of restenosis correlating well with the observed probability (r = 0.98, p = 0.0001). CONCLUSIONS: Clinical variables provide limited ability to predict definitively whether a particular patient will have restenosis. However, the current model may be used to predict the probability of restenosis, with some uncertainty, at least in well characterized patients who have already had angioplasty.

Angioplasty, Balloon, Coronary↗