Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Transfer 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 1,225 records · Page 68Linked to original sources
PubMed2003

Glucagon-like peptide-1 receptor is involved in learning and neuroprotection.

Glucagon-like peptide-1 (GLP-1) is a gut peptide that, together with its receptor, GLP-1R, is expressed in the brain. Here we show that intracerebroventricular (i.c.v.) GLP-1 and [Ser(2)]exendin(1-9) (HSEGTFTSD; homologous to a conserved domain in the glucagon/GLP-1 family) enhance associative and spatial learning through GLP-1R. [Ser(2)]exendin(1-9), but not GLP-1, is also active when administered peripherally. GLP-1R-deficient mice have a phenotype characterized by a learning deficit that is restored after hippocampal Glp1r gene transfer. In addition, rats overexpressing GLP-1R in the hippocampus show improved learning and memory. GLP-1R-deficient mice also have enhanced seizure severity and neuronal injury after kainate administration, with an intermediate phenotype in heterozygotes and phenotypic correction after Glp1r gene transfer in hippocampal somatic cells. Systemic administration of [Ser(2)]exendin(1-9) in wild-type animals prevents kainate-induced apoptosis of hippocampal neurons. Brain GLP-1R represents a promising new target for both cognitive-enhancing and neuroprotective agents.

Administration, Intranasal↗
PubMed2003

Transfer of spatial context from visual to haptic search.

Under incidental learning conditions, spatial layouts can be acquired implicitly and facilitate visual search (contextual-cueing effect). We examined whether the contextual-cueing effect is specific to the visual modality or transfers to the haptic modality. The participants performed 320 (experiment 1) or 192 (experiment 2) visual search trials based on a typical contextual-cueing paradigm, followed by haptic search trials in which half of the trials had layouts used in the previous visual search trials. The visual contextual-cueing effect was obtained in the learning phase. More importantly, the effect was transferred from visual to haptic searches; there was greater facilitation of haptic search trials when the spatial layout was the same as in the previous visual search trials, compared with trials in which the spatial layout differed from those in the visual search. This suggests the commonality of spatial memory to allocate focused attention in both visual and haptic modalities.

Adult↗
PubMed1988

Facilitative effects of cross-modality training on recovery of a conditioned avoidance response following striate cortex ablations in the rat.

Two experiments with visual decorticate rats examined the effects of cross-modality transfer on the recovery of a preoperatively learned avoidance task using high intensity light cues. In both experiments, brief postoperative cross-modality training with high intensity noise cues produced significantly better recovery of the visual avoidance response than either reversed intensity noise training or retraining with the original task. Training with a reversed intensity visual cue produced relearning deficits. In the second experiment, control conditions eliminated re-exposure to the test environment, the shock-response relationship, and compound conditioning as possible explanations for these findings. These results indicated that rats are capable of cross-modality transfer after posterior cortical lesions and that this transfer effect can facilitate recovery of behavioral function. The clinical implications of these findings are also discussed.

Animals↗
PubMed2000

Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research.

Artificial neural networks (ANNs) are biologically inspired computer programs designed to simulate the way in which the human brain processes information. ANNs gather their knowledge by detecting the patterns and relationships in data and learn (or are trained) through experience, not from programming. An ANN is formed from hundreds of single units, artificial neurons or processing elements (PE), connected with coefficients (weights), which constitute the neural structure and are organised in layers. The power of neural computations comes from connecting neurons in a network. Each PE has weighted inputs, transfer function and one output. The behavior of a neural network is determined by the transfer functions of its neurons, by the learning rule, and by the architecture itself. The weights are the adjustable parameters and, in that sense, a neural network is a parameterized system. The weighed sum of the inputs constitutes the activation of the neuron. The activation signal is passed through transfer function to produce a single output of the neuron. Transfer function introduces non-linearity to the network. During training, the inter-unit connections are optimized until the error in predictions is minimized and the network reaches the specified level of accuracy. Once the network is trained and tested it can be given new input information to predict the output. Many types of neural networks have been designed already and new ones are invented every week but all can be described by the transfer functions of their neurons, by the learning rule, and by the connection formula. ANN represents a promising modeling technique, especially for data sets having non-linear relationships which are frequently encountered in pharmaceutical processes. In terms of model specification, artificial neural networks require no knowledge of the data source but, since they often contain many weights that must be estimated, they require large training sets. In addition, ANNs can combine and incorporate both literature-based and experimental data to solve problems. The various applications of ANNs can be summarised into classification or pattern recognition, prediction and modeling. Supervised 'associating networks can be applied in pharmaceutical fields as an alternative to conventional response surface methodology. Unsupervised feature-extracting networks represent an alternative to principal component analysis. Non-adaptive unsupervised networks are able to reconstruct their patterns when presented with noisy samples and can be used for image recognition. The potential applications of ANN methodology in the pharmaceutical sciences range from interpretation of analytical data, drug and dosage form design through biopharmacy to clinical pharmacy.

Algorithms↗
PubMed2006

Order of a 'uniform random' presentation on contextual interference in a serial tracking task.

The purpose of the present study was to examine the effect of order of a 'uniform random' presentation on contextual interference in a serial tracking task. Subjects (N=48) were randomly assigned to either a random condition or a blocked condition in the acquisition phase, in which they learned three patterns of serial tracking tasks. Subjects performed retention and transfer trials either in a blocked or a 'uniform random' order. Analysis indicated, in comparisons of performance among conditions, 'uniform random' presentation in the acquisition phase did not improve learning in retention and transfer phases. The results are discussed with regard to both theoretical significance to contextual interference and the nature of 'uniform random' presentation.

Adult↗
PubMed2002

Revisiting the picture-superiority effect in symbolic comparisons: do pictures provide privileged access?

In 4 experiments, symbolic comparisons were investigated to test semantic-memory retrieval accounts espousing processing advantages for picture over word stimuli. In Experiment 1, participants judged pairs of animal names or pictures by responding to questions probing concrete or abstract attributes (texture or size, ferocity or intelligence). Per pair, attributes were salient or nonsalient concerning their prerated relevance to animals being compared. Distance (near or far) between attribute magnitudes was also varied. Pictures did not significantly speed responding relative to words across all other variables. Advantages were found forfar attribute magnitudes (i.e., the distance effect) and salient attributes. The distance effect was much less for salient than nonsalient concrete-attribute comparisons. These results were consistently found in additional experiments with increased statistical power to detect modality effects. Our findings argue against dual-coding and some common-code accounts of conceptual attribute processing, urging reexamination of the assumption that pictures confer privileged access to long-term knowledge.

Analysis of Variance↗
PubMed1991

Biological electron transfer: progress and future directions.

The rich diversity among bacterial cytochromes has played a key role in the development of our understanding of biological electron transfer. Although studies to date have allowed the elucidation of the contributions of driving force, electrostatics interactions and surface topology to electron transfer kinetics in collision-dependent reactions, much remains to be learned. Little is known about intramolecular and intracomplex electron transfer. Several factors controlling intramolecular and intracomplex electron transfer can be defined. These include driving force, the distance between redox centers, the relative orientation of prosthetic groups, the nature of the intervening media and the molecular dynamics within the electron transfer complex. However, at the present time, we have only a limited understanding of the contribution of these factors to electron transfer kinetics in biologically relevant systems. Nevertheless, a wide range of techniques are now available which should soon provide the information necessary to describe in molecular terms the mechanism of intramolecular and intracomplex electron transfer. Principal among these new approaches are site-directed mutagenesis and NMR spectroscopy.

Animals↗
PubMed2005

Lessons learned: a successful distance learning collaborative between the Department of Veterans Affairs and the Department of Defense.

Academic, business, and government organizations are increasingly looking to distance education to meet workforce learning needs. This modality differs from classroom education, however, and simply transferring content from a traditional classroom to a distance learning application can result in poor outcomes. The Department of Veterans Affairs and the Uniformed Services University of the Health Sciences Graduate School of Nursing implemented a program preparing adult nurse practitioners solely through the medium of distance education, with no university "in-residence" requirement. Lessons learned from this success can prove useful to organizations that are planning distance education initiatives.

Curriculum↗
PubMed2001

Effects of the uterine environment and neocortical ectopias upon behavior of BXSB-Yaa+ mice.

Between 40-60% of BXSB-Yaa+mice have neocortical ectopias, a genetically based brain anomaly. The presence of ectopias is known to affect several cognitive measures. A second way to affect cognition is by transferring embryos into foreign uteri. These variables were jointly investigated in three experiments. BXSB-Yaa+mouse embryos were transferred into same-strain uteri, or into uteri of hybrid mice. At birth, pups were cross-fostered to hybrid mothers or were reared by their birth mothers. When adult, the mice were given a series of behavioral tests with primary emphasis upon cognitive competence. Across all three studies, mice transferred into hybrid uteri were superior in the Morris maze and the Lashley III maze, and performed more effectively in shuttlebox avoidance learning. They were less effective in the simple water escape task, and the uterus groups did not differ in discrimination learning. Thus, development within a foreign uterus enhanced spatial learning and fear-based conditioning. Ectopic mice were superior to non-ectopics in learning the Morris maze, a finding consistent with prior research using the congenic BXSB-Yaa strain. There were Uterus x Ectopia interactions on a few measures, indicating that, under certain condition, whether the presence of ectopias is beneficial or detrimental is contingent upon the uterine environment within which the organism develops.

Animals↗
PubMed2007

The effects of cultural learning in populations of neural networks.

Population learning can be described as the iterative Darwinian process of fitness-based selection and genetic transfer of information leading to populations of higher fitness and is often simulated using genetic algorithms. Cultural learning describes the process of information transfer between individuals in a population through non-genetic means. Cultural learning has been simulated by combining genetic algorithms and neural networks using a teacher-pupil scenario where highly fit individuals are selected as teachers and instruct the next generation. By examining the innate fitness of a population (i.e., the fitness of the population measured before any cultural learning takes place), it is possible to examine the effects of cultural learning on the population's genetic makeup. Our model explores the effect of cultural learning on a population and employs three benchmark sequential decision tasks as the evolutionary task for the population: connect-four, tic-tac-toe, and blackjack. Experiments are conducted with populations employing population learning alone and populations combining population and cultural learning. The article presents results showing the gradual transfer of knowledge from genes to the cultural process, illustrated by the simultaneous decrease in the population's innate fitness and the increase of its acquired fitness measured after learning takes place.

Algorithms↗
PubMed2005

Effector-dependent and response location learning of probabilistic sequences in serial reaction time tasks.

We investigated the contributions of the sequences of effectors and response locations to probabilistic sequence learning in the serial reaction time task by means of bimanual transfer. Participants, trained with the dominant hand, were either required to maintain responding with the dominant hand after transfer or to switch to the nondominant hand. For both groups, half of the participants were transferred to the originally trained sequence, whereas the other half was transferred to a mirror-ordered sequence. This way, the sequence of effectors varied independently of the sequence of response locations. Sequence learning was assessed with probabilistic sequences, composed of either first-order or second-order probabilities. Transfer of sequence knowledge to the nondominant hand was equally good for the originally trained sequence as for the mirrored sequence. This suggests that probabilistic sequence learning can be based on either the sequence of effectors or response locations. However, when participants responded with the dominant hand to a mirrored sequence, transfer performance was disturbed. This indicates that changing both the sequences of effectors and response locations has a disruptive effect on the learning performance.

Adult↗
PubMed1994

Interocular transfer in the split-brain rat.

After the 1991 development of a method to transect the optic chiasm in the rat, the authors investigated interocular transfer in split-chiasm rats with corpus callosum section and callosum intact. Rats were monocularly trained on an orientation discrimination, reversed with the other eye, and retrained with the 1st eye on the 2nd-eye problem or reversal. In split-chiasm rats, the 2nd-eye S+ (correct stimulus) showed strong transfer (55% savings) to the 1st eye. S- (incorrect stimulus) did not; reversal and transfer groups differed significantly. Split-brain rats retrained with the 1st eye on the 2nd eye S+ required no more trials than original learning, as if they had not experienced this problem. Split-brain rats retested with the original S+ showed modest (19%) savings. The findings establish interhemispheric communication via the callosum and suggest, as in findings of cat and monkey studies, the disruption of transfer by callosal section.

Animals↗
PubMed2002

Individual differences in spatial memory among aged rats are related to hippocampal PKCgamma immunoreactivity.

We reported previously that the extent of spatial memory impairment among aged rats was correlated positively with levels of protein kinase Cgamma in hippocampal homogenates measured by quantitative Western blotting (Colombo et al., 1997). In the current study, immunocytochemistry was used to test whether the relationship between elevated PKC-gamma and memory impairment among aged rats could be localized further within regions of the hippocampus. Six- and 24-month-old male Long-Evans rats were first trained in the water maze on a standard place-learning task and then trained 2 weeks later on a transfer task designed for rapid acquisition. In comparison with young rats, aged rats with impaired spatial memory had increased PKCgamma-immunoreactivity (PKCgamma-ir) in CA1 of the hippocampus, but not the dentate gyrus. In addition, PKCgamma-ir in CA1 was correlated positively with spatial memory impairment among aged rats on the standard place-learning and the transfer training tasks. The current results are consistent with our previous report of PKCgamma in hippocampal homogenates, and show further that the relationships between PKCgamma-ir and memory impairments among aged rats are most evident in area CA1. Thus age-related impairments of spatial memory, as well as deficits in the flexible use of previously acquired information, may result from dysregulation of PKCgamma.

Aging↗
PubMed1979

Congenital ocular motor apraxia. A possible disconnection syndrome.

In 1952, Cogan introduced the term "congenital ocular motor apraxia" (COA) to describe an abnormality of eye movements characterized by absent or defective voluntary horizontal gaze. Since his original description, there have been few subsequent reports of this disorder. A ten-year review of clinical records from the University of Wisconsin Hospitals disclosed eight patients with COA. In two patients subjected to neuroradiologic testing, agenesis of the corpus callosum was detected. Voluntary horizontal gaze appears to be a learned phenomenon, and defective interhemispheric transfer of visual information may be important in the pathogenesis of COA.

Agenesis of Corpus Callosum↗
PubMed2001

Follow-up in train-the-trainer continuing medical education events.

BACKGROUND: The purpose of train-the-trainer (TTT) programs within the context of continuing medical education (CME) is to help facilitators acquire and/or enhance their skills at leading CME sessions. The provision of follow-up is one feature of successful CME workshops over which CME providers have some control. Follow-up is defined as any encounter between participants and workshop leaders, following an initial workshop or other development session, and is designed to enhance, maintain, reinforce, transfer, extend, or support the learning from the original workshop. In this article, we elaborate on the use of audio teleconferences to provide follow-up for a TTT workshop in Saskatchewan, a largely rural province in western Canada. METHODS: The teleconferences began 6 weeks after the workshop and were held at approximately 6-week intervals, with five conference calls in total. Each lasted about 45 minutes. Participants were interviewed to determine their view of the value of the teleconferences. RESULTS: Participants reported learning from the teleconferences and feeling more prepared to conduct CME sessions due to their participation in the teleconferences. Participants missed teleconferences only for extenuating circumstances (e.g., emergency deliveries). FINDINGS: We have found that audio teleconferences allow for and encourage professional discussion that is crucial to changing practices. They are an effective way to incorporate follow-up to TTT workshops when participants travel great distances to attend.

Education, Medical, Continuing↗