SEARCH · Search PubMed

Results for “sparse 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.

SourceResearch connectionFrom yearThrough year

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

At least 181 records · Page 10Linked to original sources

PubMed2004

The impact of lexical factors on children's word-finding errors.

This retrospective, exploratory investigation examined the types of target words that 30 children with word-finding difficulties (aged 8 to 12 years) had difficulty naming and the types of errors they made on these words. Words were studied with reference to lexical factors that might influence naming performance: word frequency, age of acquisition, familiarity, and lexical neighborhood. Findings indicated that neighborhood density predicted word-finding success, and target word substitutions and error patterns manifested were affected by the lexical factors under study. Students tended to produce substitutions that were higher in frequency, learned earlier, and that resided in neighborhoods of greater density and higher frequency than the target word. Lexical factors also influenced children's error patterns. Neighborhood density predicted form-related errors: Children produced more blocked errors on words from sparse neighborhoods. Word frequency and neighborhood frequency predicted form-segment-related errors as phonologic errors occurred on rare words and words whose neighbors contained lower frequency, uncommon phonological patterns. This important first step in the examination of how lexical factors have an impact on word-finding errors in children suggests that different types of words are more likely to result in failures of lexical access at different stages of processing. Theoretical and practical implications of these preliminary findings are discussed.

Analysis of Variance↗

PubMed1988

Kabuki make-up (Niikawa-Kuroki) syndrome: a study of 62 patients.

These 62 patients with the Kabuki make-up syndrome (KMS) were collected in a collaborative study among 33 institutions and analyzed clinically, cytogenetically, and epidemiologically to delineate the phenotypic spectrum of KMS and to learn about its cause. Among various manifestations observed, most patients had the following five cardinal manifestations: 1) a peculiar face (100%) characterized by eversion of the lower lateral eyelid; arched eyebrows, with sparse or dispersed lateral one-third; a depressed nasal tip; and prominent ears; 2) skeletal anomalies (92%), including brachydactyly V and a deformed spinal column, with or without sagittal cleft vertebrae; 3) dermatoglyphic abnormalities (93%), including increased digital ulnar loop and hypothenar loop patterns, absence of the digital triradius c and/or d, and presence of fingertip pads; 4) mild to moderate mental retardation (92%); and 5) postnatal growth deficiency (83%). Thus the core of the phenotypic spectrum of KMS is rather narrow and clearly defined. Many other inconsistent anomalies were observed. Important among them were early breast development in infant girls (23%), and congenital heart defects (31%), such as a single ventricle with a common atrium, ventricular septal defect, atrial septal defect, tetralogy of Fallot, coarctation of aorta, patent ductus arteriosus, aneurysm of aorta, transposition of great vessels, and right bundle branch block. Of the 62 KMS patients, 58 were Japanese, an indication that the syndrome is fairly common in Japan. It was estimated that its prevalence in Japanese newborn infants is 1/32,000. All the KMS cases in this study were sporadic, the sex ratio was even, there was no correlation with birth order, the consanguinity rate among the parents was not high, and no incriminated agent was found that was taken by the mothers during early pregnancy. Three of the 62 patients had a Y chromosome abnormality involving a possible common breakpoint (Yp11.2). This could indicate another possibility, i.e., that the KMS gene is on Yp11.2 and that the disease is pseudoautosomal dominant. These findings are compatible with an autosomal dominant disorder in which every patient represents a fresh mutation. The mutation rate was calculated at 15.6 X 10(6).

Abnormalities, Multiple↗

PubMed2005

The HVC microcircuit: the synaptic basis for interactions between song motor and vocal plasticity pathways.

Synaptic interactions between telencephalic neurons innervating descending motor or basal ganglia pathways are essential in the learning, planning, and execution of complex movements. Synaptic interactions within the songbird telencephalic nucleus HVC are implicated in motor and auditory activity associated with learned vocalizations. HVC contains projection neurons (PNs) (HVC(RA)) that innervate song premotor areas, other PNs (HVC(X)) that innervate a basal ganglia pathway necessary for vocal plasticity, and interneurons (HVC(INT)). During singing, HVC(RA) fire in temporally sparse bursts, possibly because of HVC(INT)-HVC(RA) interactions, and a corollary discharge can be detected in the basal ganglia pathway, likely because of synaptic transmission from HVC(RA) to HVC(X) cells. During song playback, local interactions, including inhibition onto HVC(X) cells, shape highly selective responses that distinguish HVC from its auditory afferents. To better understand the synaptic substrate for the motor and auditory properties of HVC, we made intracellular recordings from pairs of HVC neurons in adult male zebra finch brain slices and used spike-triggered averages to assess synaptic connectivity. A major synaptic interaction between the PNs was a disynaptic inhibition from HVC(RA) to HVC(X), which could link song motor signals in the two outputs of HVC and account for some of the song playback-evoked inhibition in HVC(X) cells. Furthermore, single interneurons made divergent connections onto PNs of both types, and either PN type could form reciprocal connections with interneurons. In these two regards, the synaptic architecture of HVC resembles that described in some pattern-generating networks, underscoring features likely to be important to singing and song learning.

2-Amino-5-phosphonovalerate↗

PubMed1991

What might echography learn from image science?

We review the current state of knowledge of the processes by which the information content of ultrasonic pulse-echo images is transferred to an observer, to the point of contributing to diagnostic judgments. As systematic knowledge in this specific field is rather sparse, we present relevant information and techniques derived from other areas of image science, both medical and otherwise. Quantitative measures both of the information content of ultrasonic and other images and of their characteristic noise content are first considered. An account is then given of the relevant aspects of human visual psychophysics, with particular reference to perception of contrast and detail, image texture, movement and colour, again with emphasis on documenting quantitative aspects of such behaviour. Against this background, we consider the efficiency, in current practice, of image information transfer to a human observer, how and to what extent this could be improved by changes in practice and, in particular, in what situations substantial innovations in machine processing of image data would be expected to improve human performance. It is suggested that several problems in the field may provide a worthwhile and challenging scope for future research.

Humans↗

PubMed2002

Evidence for longitudinal ambulatory care rotations: a review of the literature.

PURPOSE: Block ambulatory rotations and longitudinal ambulatory care experiences are now common in U.S. medical schools, but little is known about their efficacy. Through a structured review of the medical literature from 1966 through March 2000, the authors summarize the characteristics of, the evidence for, and the evaluation of longitudinal ambulatory care rotations. METHOD: The authors searched Medline using the terms "outpatients," "continuity of patient care," "ambulatory care," "mentors," "preceptorship," "graduate medical education," "curriculum," and "clinical clerkship" cross-matched to "medical students" and "internship and residency" for literature published from 1966 through March 2000. They narrowed the list to only articles containing empirical outcome data focusing on medical students' experiences in longitudinal ambulatory care rotations. Each study was scored to assess its strengths and weaknesses. RESULTS: Seven articles met the search criteria. The articles identified the benefits of longitudinal ambulatory care experiences, including developing effective patient interactions and understanding chronic diseases. There were little or no differences in the students' overall knowledge acquisition when those with longitudinal experiences were compared with those in block rotations. DISCUSSION: Although longitudinal ambulatory care experiences are now common in medical schools, evidence supporting their widespread implementation is sparse. Few studies employ rigorous methods to evaluate educational outcomes. Research to identify benefits and costs, improve the quality and consistency of the students' experiences, and develop other innovative ways of teaching and learning ambulatory care is needed.

Ambulatory Care↗

PubMed2005

Levetiracetam in the treatment of idiopathic generalized epilepsies.

Since its introduction into clinical practice in 1999, levetiracetam, the S enantiomer of piracetam, has rapidly found a secure place, initially in the therapy of partial onset seizures and subsequently in the treatment of idiopathic generalized epilepsies (IGE). It has many of the properties of an "ideal" antiepileptic drug, including rapid absorption, linear pharmokinetics, and sparse drug interactions. Tolerabiliy is generally excellent in both adults and children, although tiredness is a common dose-limiting adverse effect. Occasionally the drug can precipitate behavioral abnormalities, especially in patients with learning disability. There is a wide safety margin in overdose. In common with most antiepileptic drugs its mode of action remains uncertain. Levetiracetam binds to a specific site in the brain, influences intracellular calcium currents and reverses negative allosteric modulators of GABA- and glycine-gated currents in vitro. Its effectiveness has been demonstrated in animal models of epilepsy and in clinical trials of partial onset and IGE. Treatment of IGEs may be straightforward, with many patients demonstrating an excellent and robust response to valproate monotherapy. However, there remains a significant minority of patients for whom valproate is unsuitable, including those who experience unacceptable adverse effects (e.g., weight gain or hair loss) and women of childbearing age in whom the teratogenic potential of valproate is unacceptable. Therapeutic response to lamotrigine in this group is often disappointing, and many clinicians now are turning to the choice of levetiracetam. Efficacy in generalized tonic-clonic seizures and myoclonus is usually apparent and some patients experience improvement in typical absences. Experience of combinations of levetiracetam with other antiepileptic drugs is limited in IGE and the responses are largely anecdotal. In our hands, patients with refractory IGEs may respond to combinations of levetiracetam with valproate, lamotrigine, and phenobarbital, and adverse effects when they occur are usually limited to tiredness. Levetiracetam does not interact with the oral contraceptive pill, simplifying treatment in women of childbearing age. Although animal data look encouraging, questions over levetiracetam's teratogenic potential and overall safety in pregnancy will remain for many years to come.

Adult↗

PubMed2001

The limits of counting accuracy in distributed neural representations.

Learning about a causal or statistical association depends on comparing frequencies of joint occurrence with frequencies expected from separate occurrences, and to do this, events must somehow be counted. Physiological mechanisms can easily generate the necessary measures if there is a direct, one-to-one relationship between significant events and neural activity, but if the events are represented across cell populations in a distributed manner, the counting of one event will be interfered with by the occurrence of others. Although the mean interference can be allowed for, there is inevitably an increase in the variance of frequency estimates that results in the need for extra data to achieve reliable learning. This lowering of statistical efficiency (Fisher, 1925) is calculated as the ratio of the minimum to actual variance of the estimates. We define two neural models, based on presynaptic and Hebbian synaptic modification, and explore the effects of sparse coding and the relative frequencies of events on the efficiency of frequency estimates. High counting efficiency must be a desirable feature of biological representations, but the results show that the number of events that can be counted simultaneously with 50% efficiency is fewer than the number of cells or 0.1-0.25 of the number of synapses (on the two models) - many fewer than can be unambiguously represented. Direct representations would lead to greater counting efficiency, but distributed representations have the versatility of detecting and counting many unforeseen or rare events. Efficient counting of rare but important events requires that they engage more active cells than common or unimportant ones. The results suggest reasons that representations in the cerebral cortex appear to use extravagant numbers of cells and modular organization, and they emphasize the importance of neuronal trigger features and the phenomena of habituation and attention.

Animals↗

PubMed2006

Orthogonality is not a panacea: backpropagation and "catastrophic interference".

Connectionist models with the backpropagation learning rule are said to exhibit catastrophic interference (or forgetting) with sequential training. Subsequent works showed that interference can be reduced by using orthogonal inputs. This study investigated, with a more rigorous assessment method, whether all orthogonal inputs lead to comparable extent of interference using three coding schemes. The results revealed large differences between the coding schemes. With larger networks, dense inputs led to severer interference compared with sparse inputs. With smaller networks, all the three schemes led to comparable extent of interference. Therefore, this study proved that not all the orthogonal inputs cause the same extent of interference, and that severity of interference depends on the interaction of the input coding scheme and the network size.

Humans↗

PubMed2002

The development of decision-making.

PURPOSE: To describe what is currently known and not known about the development of decision-making skills during adolescence. METHODS: The author provides a definition of competent decision-making, gives a brief overview of the literature on the development of this competence, and describes the kinds of studies that should be conducted in the near future. RESULTS AND CONCLUSIONS: Although the literature is still too sparse to be conclusive at present, the literature as a whole suggests that adolescents may have less decision-making competence than adults in certain areas (i.e., advice-seeking, evaluation processes, adaptive goal-setting, and learning) but may have similar levels of competence in other areas (i.e., knowledge of options in familiar areas, response to certain moderating factors, and making choices in a number of areas). Moreover, age differences in deliberative aspects of competence have not yet been linked to age differences in the tendency to pursue good options, although one recent study did find a possible link between a post-decisional process (i.e., learning from decisions) and age differences in the selection of good options. Additional studies are needed to firm up the tentative conclusions that can be drawn from the extant research and determine which kinds of interventions improve decision-making in adolescents and which do not.

Adolescent↗

PubMed2004

ESPD: a pattern detection model underlying gene expression profiles.

MOTIVATION: DNA arrays permit rapid, large-scale screening for patterns of gene expression and simultaneously yield the expression levels of thousands of genes for samples. The number of samples is usually limited, and such datasets are very sparse in high-dimensional gene space. Furthermore, most of the genes collected may not necessarily be of interest and uncertainty about which genes are relevant makes it difficult to construct an informative gene space. Unsupervised empirical sample pattern discovery and informative genes identification of such sparse high-dimensional datasets present interesting but challenging problems. RESULTS: A new model called empirical sample pattern detection (ESPD) is proposed to delineate pattern quality with informative genes. By integrating statistical metrics, data mining and machine learning techniques, this model dynamically measures and manipulates the relationship between samples and genes while conducting an iterative detection of informative space and the empirical pattern. The performance of the proposed method with various array datasets is illustrated.

Algorithms↗

PubMed2002

A review of current practice for boys undergoing hypospadias repair: from pre-operative work up to removal of dressing post-surgery.

The aim of this article was to examine an area from dinical practice, in this instance, paediatric urology and to look at the information needs of families prior to their young sons' having surgery to correct a hypospadias. Involving families and promoting informed decision-making with families prior to hypospadias repair for their child is important. To be able to facilitate real involvement the information has to be shared and parents and families given adequate support and opportunity to seek answers to questions. Providing an environment to undertake this, to then learn from the families for the future and to support the development of a National Support group, www.hypospadias.co.uk have been outcomes from setting up a pre-admission clinic. Literature in respect of after-care is very sparse. It became evident from clinical practice that the rationale for applying a dressing postoperatively needed to be explored. Furthermore limited evidence about dressing removal, coupled with the anxiety and upset this caused parents, identified the need to consider two issues (i) a better method of removal of dressings and (ii) the effects on pain scores. A blinded randomized controlled feasibility study was developed and implemented, with completion due at the end of 2001.

Aftercare↗

PubMed2004

Species-typical songs in white-crowned sparrows tutored with only phrase pairs.

Modern theories of learned vocal behaviours, such as human speech and singing in songbirds, posit that acoustic communication signals are reproduced from memory, using auditory feedback. The nature of these memories, however, is unclear. Here we propose and test a model for how complex song structure can emerge from sparse sequence information acquired during tutoring. In this conceptual model, a population of combination-sensitive (phrase-pair) detectors is shaped by early exposure to song and serves as the minimal representation of the template necessary for generating complete song. As predicted by the model, birds that were tutored with only pairs of normally adjacent song phrases were able to assemble full songs in which phrases were placed in the correct order; birds that were tutored with reverse-ordered phrase pairs sang songs with reversed phrase order. Birds that were tutored with all song phrases, but presented singly, failed to produce normal, full songs. These findings provide the first evidence for a minimal requirement of sequence information in the acoustic model that can give rise to correct song structure.

Acoustic Stimulation↗

PubMed2006

Sparse Bayesian kernel survival analysis for modeling the growth domain of microbial pathogens.

Survival analysis is a branch of statistics concerned with the time elapsing before "failure," with diverse applications in medical statistics and the analysis of the reliability of electrical or mechanical components. We introduce a parametric accelerated life survival analysis model based on kernel learning methods that, at least in principal, is able to learn arbitrary dependencies between a vector of explanatory variables and the scale of the distribution of survival times. The proposed kernel survival analysis method is then used to model the growth domain of Clostridium botulinum, the food processing and storage conditions permitting the growth of this foodborne microbial pathogen, leading to the production of the neurotoxin responsible for botulism. A Bayesian training procedure, based on the evidence framework, is used for model selection and to provide a credible interval on model predictions. The kernel survival analysis models are found to be more accurate than models based on more traditional survival analysis techniques but also suggest a risk assessment of the foodborne botulism hazard would benefit from the collection of additional data.

Artificial Intelligence↗

PubMed1997

Sparsely Connected, Hebbian Networks with Strikingly Large Storage Capacities.

Conspicuous problems confront the use of fully-connected networks (F-nets) in the construction of realistic partial models of biological memory. These problems include the high synaptic densities of F-nets, and the low information storage capacities of F-nets with simple, biologically realistic learning rules. Most auto-associative networks constructed with low connectivities have employed random projections of path length 1. Projective networks (P-nets) are nonrandom, multilayer networks which achieve extremely low connectivities by linking all neurons in the same layer through paths of length 2. In this paper we derive a lower bound on the storage capacities of a class of simple, two-layer P-nets with binary Hebbian synapses. Given a 1% tolerance for spurious neurons, we find that the P-net with 1000 synapses per neuron (2 x 10(6) neurons) will store more than 1.5 x 10(6) training vectors with 20 active neurons per vector (0.25 bits per synapse). Copyright 1997 Elsevier Science Ltd.

Journal Article↗

PubMed1982

Histologic and fine-structural anatomy of the adrenorenal portal vessels in the dog.

Reflex vasoconstriction that occurs in the kidney of the dog can be the result of activation of catecholamine flow reaching the kidney directly from the adrenal medulla via adrenorenal portal vessels. This study was undertaken to learn more about the histologic and fine-structural anatomy of these unusual vessels. We found that they are thin-walled vessels (200-500 micrometers) which are accompanied by many capillaries. These portal vessels have distinctive valves and have sparse, longitudinally arranged musculature which is denser at the base of the valves. The valves in the portal vessels are arranged to permit flow in the direction from the adrenal gland toward the kidney. Smooth muscle at the base of the valve is in close proximity to non-myelinated nerve fibers, suggesting that the opening of these valves may in part be neurally mediated. These unique anatomical features distinguish the adrenorenal portal vessels from other blood vessels and likely have functional importance in control of the adrenorenal circulation.

Adrenal Medulla↗

PubMed2000

Impaired cognitive performance in ornithine transcarbamylase-deficient mice on arginine-free diet.

Sparse-fur (spf) mice are a model for the congenital deficiency of ornithine transcarbamylase (OTC), the most common inborn error of urea synthesis in man. In this study, performance of clinically stable spf and control mice (8-10-weeks-old) on two learning tests was assessed under normal Arg(+) or arginine-free Arg(-) diet conditions. Used as an indicator of the metabolic status of the animals, plasma ammonia concentrations were significantly higher in spf than in controls on normal diet, and increased even more during the Arg(-) diet episode. Behaviourally, we found no difference in passive avoidance learning between control and spf mice on Arg(+) diet, whereas in spf mice receiving Arg(-) diet during training, retention performance was significantly reduced. In the hidden-platform water maze, spf mice on Arg(+) diet only showed decreased swimming velocity compared to controls. In mice on Arg(-) diet during the first week of acquisition training, performance on acquisition and retention (probe) trials showed that spf mice experienced more difficulties in actually locating the platform. Visible-platform control experiments only showed a reduction in swimming velocity in spf mice on either diet. We conclude that cognitive performance is impaired in spf mice as a consequence of Arg(-) diet-induced neurochemical alterations.

Ammonia↗

PubMed2003

A Turkish case of subcortical/subependymal heterotopia associated with corpus callosum dysgenesis, craniofacial dysmorphism, severe eye abnormalities, and growth-mental retardation.

The patient is a 12-year-old boy with a history of learning disability, growth retardation, and strabismus. Weight, height and head circumference were below the 3rd percentile. A café-au-lait spot, 1x1 cm a diameter, on the back region and pectus excavatum deformity were diagnosed. He had facial asymmetry, a broad nose, sparse eyebrows and eyelashes, a rudimentary frontal sinus, deviation of the nasal septum, and bilateral small maxillary bones. The left orbital fossa was also mildly rudimentary. On eye examination the movements of the left globe to the upward and lateral side were limited and internal strabismus was noted at this side. Visual acuity was 1/10, bilaterally. Bilateral choroid coloboma, glaucoma, vertical and horizontal nystagmus were diagnosed. Fundoscopic examination revealed bilateral optic atrophy and macular and paramacular granulation tissues on the left side. Intelligence quotient was 46. Electroencephalography revealed bilateral frontal slow-wave activity. Visual evoked potential revealed prolonged p100 wave latencies bilaterally. Magnetic resonance imaging of the brain demonstrated corpus callosum dysgenesis, bilateral subcortical heterotopia in the frontal lobes and subependymal heterotopia in the posterior horn of the left ventricle. Chromosomal analysis revealed a normal male karyotype, 46, XY. Although several cases of heterotopia in association with mental retardation, craniofacial dysmorphism, cerebral, and eye abnormalities have been described the combination of abnormalities diagnosed in our case has not previously been reported. We hypothesize that the combination of subcortical/subependymal heterotopia, corpus callosum dysgenesis, craniofacial dysmorphism, severe eye abnormalities, and growth-mental retardation may be a new syndrome.

Abnormalities, Multiple↗

PubMed1998

A recurrent neural network for modelling dynamical systems.

We introduce a recurrent network architecture for modelling a general class of dynamical systems. The network is intended for modelling real-world processes in which empirical measurements of the external and state variables are obtained at discrete time points. The model can learn from multiple temporal patterns, which may evolve on different timescales and be sampled at non-uniform time intervals. We demonstrate the application of the model to a synthetic problem in which target data are only provided at the final time step. Despite the sparseness of the training data, the network is able not only to make good predictions at the final time step for temporal processes unseen in training, but also to reproduce the sequence of the state variables at earlier times. Moreover, we show how the network can infer the existence and role of state variables for which no target information is provided. The ability of the model to cope with sparse data is likely to be useful in a number of applications, including, in particular, the modelling of metal forging.

Models, Theoretical↗