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 1,117 records · Page 62Linked to original sources

Adult-infant food transfer in common marmosets: an experimental study.

Infant marmosets and tamarins obtain solid food items from adults during and after the time of weaning. In addition to providing nutrients, food transfers may provide infants with the opportunity to learn about diet. The aim of this study was to investigate patterns of begging and food transfer in captive groups of common marmosets during tests with 1) palatable novel food, 2) unpalatable novel food, and 3) familiar food. Infants may gain an opportunity to learn about diet breadth by preferentially begging for novel rather than familiar items, while adults may facilitate infant learning by differing in their willingness to transfer food depending upon the food's novelty or palatability. In this study, infants exhibited more interest and begging with novel food items than with familiar ones. However, food transfers were not more likely to occur with novel foods than with familiar ones, and in fact adults were more likely to refuse infant begging attempts with novel foods. The palatability of the food also did not influence the rates of refusal and transfer: adults transferred palatable and unpalatable novel foods with similar frequencies. These results suggest that infant marmosets gain an opportunity to learn about diet breadth by begging for novel foods, but adults do not preferentially transfer novel or palatable food items.

Analysis of Variance↗

Training a perceptron in a discrete weight space.

Learning in a perceptron having a discrete weight space, where each weight can take 2L+1 different values, is examined analytically and numerically. The learning algorithm is based on the training of the continuous perceptron and prediction following the clipped weights. The learning is described by a new set of order parameters, composed of the overlaps between the teacher and the continuous/clipped students. Different scenarios are examined, among them on-line learning with discrete and continuous transfer functions. The generalization error of the clipped weights decays asymptotically as exp(-Kalpha(2)) in the case of on-line learning with binary activation functions and exp(-e(|lambda|alpha)) in the case of on-line learning with continuous one, where alpha is the number of examples divided by N, the size of the input vector and K is a positive constant. For finite N and L, perfect agreement between the discrete student and the teacher is obtained for alpha~Lsqrt[ln(NL)]. A crossover to the generalization error approximately 1/alpha, characterizing continuous weights with binary output, is obtained for synaptic depth L>O(sqrt[N]).

Journal Article↗

Independent processing of the temporal and ordinal structure of movement sequences.

We investigated if the temporal and ordinal structures of sequences can be represented and learned independently. In Experiment 1, subjects learned three rhythmic sequences of key presses with the right index finger: Combined consisted of nine key presses with a corresponding temporal structure of eight intervals; Temporal had the temporal structure of Combined but was performed on one key; Ordinal had the ordinal structure of Combined but an isochronous rhythm. Subjects were divided into two groups. Group 1 first learned Combined, then Temporal and Ordinal; Group 2 first learned Temporal and Ordinal, then Combined. Strong transfer effects were seen in both groups. In Group 1, having learned combined facilitated the learning of the temporal (Temporal) or ordinal (Ordinal) sequence alone; in Group 2, having learned Temporal and Ordinal facilitated the learning of Combined, where the two are combined. This supports that subjects had formed independent temporal and ordinal representations. In Experiment 2, we investigated if these can be learned independently. Subjects repeatedly reproduced sequences with fixed temporal and random ordinal structure; random temporal and fixed ordinal structure; and random temporal and ordinal structures. Temporal and ordinal learning was seen only in the first and second sequences, respectively. In summary, we provide evidence for the existence of independent systems for learning and representation of ordinal and temporal sequences and for implicit learning of temporal sequences. This may be important for fast learning and flexibility in motor control.

Adult↗

Age-related changes in barrier function in mouse brain I. Accelerated age-related increase of brain transfer of serum albumin in accelerated senescence prone SAM-P/8 mice with deficits in learning and memory.

The time course of brain accumulation of radiolabelled human serum albumin ((125)I-HSA) injected intravenously and the transfer of (125)I-HSA from blood to brain were evaluated in DDD mice using a double isotope technique. The brain accumulation of (125)I-HSA at 3 and 9 h but not at 24 h postinjection and the brain transfer rates were significantly higher in 22-month-old DDD mice than in 4-month-old ones. The brain transfer rates of (125)I-HSA were measured also in senescence accelerated prone mice (SAM-P/8) with age-related deficits in learning and memory, and in senescence accelerated resistant mice (SAM-R/I) without these deficits. The brain transfer rates were significantly higher in 13-month-old SAM-P/8 and 22-month-old SAM-R/1 than in 3-month-old mice of the same strains, respectively. The mean brain transfer rates in five regions observed in 22-month-old DDD mice, 22-month-old SAM-R/1 and 13-month-old SAM-P/8 increased by 31%, 41% and 51% compared with corresponding values in 3- or 4-month-old mice of the same strains. DDD mice and SAM-R/1 mice with normal characteristics of aging showed similar age-related significant changes in brain transfer rates. Age-related increase in the brain transfer rate was manifested at the youngest age in SAM-P/8 among the three strains examined. These findings show that the transfer of human serum albumin into the mouse brain increases with aging and suggest that the barrier function in the mouse brain against macromolecules changes with aging.

Journal Article↗

Training in virtual environments: transfer to real world tasks and equivalence to real task training.

Virtual environments (VEs) are extensively used in training but there have been few rigorous scientific investigations of whether and how skills learned in a VE are transferred to the real world. This research aimed to measure and evaluate what is transferring from training a simple sensorimotor task in a VE to real world performance. In experiment 1, real world performances after virtual training, real training and no training were compared. Virtual and real training resulted in equivalent levels of post-training performance, both of which significantly exceeded task performance without training. Experiments 2 and 3 investigated whether virtual and real trained real world performances differed in their susceptibility to cognitive and motor interfering tasks (experiment 2) and in terms of spare attentional capacity to respond to stimuli and instructions which were not directly related to the task (experiment 3). The only significant difference found was that real task performance after training in a VE was less affected by concurrently performed interference tasks than was real task performance after training on the real task. This finding is discussed in terms of the cognitive load characteristics of virtual training. Virtual training therefore resulted in equivalent or even better real world performance than real training in this simple sensorimotor task, but this finding may not apply to other training tasks. Future research should be directed towards establishing a comprehensive knowledge of what is being transferred to real world performance in other tasks currently being trained in VEs and investigating the equivalence of virtual and real trained performances in these situations.

Acoustic Stimulation↗

Judging a book by its cover: interpretative effects of content on problem-solving transfer.

We examine how cover stories of isomorphic problems affect transfer. Existing models posit that people retain content in problem representations and that similarities and differences between the "undeleted" cover stories might interfere with recognition of structural similarities. We propose that cover stories can affect transfer in another way--by inducing semantic knowledge that modifies problem structures. Two experiments examined how people represent and solve permutation problems dealing with random assignment of elements from one set to elements from another set. Although the problems were structurally isomorphic, cover stories involving different pairs of element sets led subjects to abstract different "interpreted structures." Problems involving objects and people (e.g., prizes and students) led subjects to abstract an asymmetric structure ("get") and problems involving similar sets of people (e.g., doctors and doctors) led subjects to abstract a symmetric structure ("pair"). Transfer was mediated by similarities and differences between the interpreted structures of the learned and the novel problems.

Adult↗

[Learning strategies of hippocampectomized rats on an elevated maze].

Hippocampectomized (N = 19) and control (N = 18) rats were trained on an elevated T-maze in which place, cue and response strategies employed could be separated from one another. For nine hippocampals and nine controls, a place task was followed by a cue task (place-cue condition) while the order was reversed for the remaining subjects (cue-place condition). The effects of hippocampal lesions on acquisition and transfer of place and cue learning were assessed in terms of correct responses and strategies used by the rats. In both training conditions, performance of hippocampals was significantly impaired for the place task but not for the cue task. Hippocampals used a cue strategy in the cue task but employed a response strategy in the place task. Controls, on the other hand, used a place strategy in the place task and a cue strategy in the cue task. These results suggest that rats with hippocampal lesions are imparied in their ability to employ a place strategy with the result that other strategies are adopted in a place task situation.

Animals↗

Extrapolation: the sine qua non for abstraction in function learning.

Abstraction was investigated by examining extrapolation behavior in a function-learning task. During training, participants associated stimulus and response magnitudes (in the form of horizontal bar lengths) that covaried according to a linear, exponential, or quadratic function. After training, novel stimulus magnitudes were presented as tests of extrapolation and interpolation. Participants extrapolated well beyond the range of learned responses, and their responses captured the general shape of the assigned functions, with some systematic deviations. Notable individual differences were observed, particularly in the quadratic condition. The number of unique stimulus-response pairs given during training (i.e., density) was also manipulated but did not affect training or transfer performance. Two rule-learning models, an associative-learning model, and a new hybrid model with associative learning and rule-based responding (extrapolation-association model [EXAM]) were evaluated with respect to the transfer data. EXAM best approximated the overall pattern of extrapolation performance.

Humans↗

Teaching critical thinking for transfer across domains. Dispositions, skills, structure training, and metacognitive monitoring.

Advances in technology and changes in necessary workplace skills have made the ability to think critically more important than ever before, yet there is ample evidence that many adults consistently engage in flawed thinking. Numerous studies have shown that critical thinking, defined as the deliberate use of skills and strategies that increase the probability of a desirable outcome, can be learned in ways that promote transfer to novel contexts. A 4-part empirically based model is proposed to guide teaching and learning for critical thinking: (a) a dispositional component to prepare learners for effortful cognitive work, (b) instruction in the skills of critical thinking, (c) training in the structural aspects of problems and arguments to promote transcontextual transfer of critical-thinking skills, and (d) a metacognitive component that includes checking for accuracy and monitoring progress toward the goal.

Adult↗

Piracetam-induced facilitation of interhemispheric transfer of visual information in rats.

The effect of Piracetam (UCB 6215, 2-pyrrolidoneacetamide) on learning mediated by transcommissural information flow was studied in hooded rats. Acquisition of monocular pattern discrimination was faster in drug-treated rats (100 mg/kg, 30 min before training) than in untreated controls. Subsequent relearning with one hemisphere functionally eliminated by cortical spreading depression showed that the strength of the primary engram formed under Piracetam in the hemisphere contralateral to the trained eye remained unaffected but that the secondary trace (in the ipsilateral hemisphere) was considerably improved and almost equalled the primary one (savings increased from 20-30% to 50-60%). Learning with uncrossed optic fibers was unaffected by the drug. Interhemispheric transfer of lateralized visual engrams acquired during functional hemidecortication was facilitated by Piracetam administration preceding the five transfer trials performed with the untrained eye open (imperative transfer). Piracetam was ineffective when the trained eye was open during transfer trials (facultative transfer). After a visual engram had been lateralized by 5 days of monocular overtraining, Piracetam facilitated formation of the secondary engram induced by 3 interocular transfer trials. It is concluded that Piracetam enhances transcommissural encoding mechanisms activated in the initial stage of monocular learning and in some forms of interhemispheric transfer, but does not affect the transcommissural readout. This effect is interpreted as a special case of the Piracetam-induced facilitation of the phylogenetically old mechanisms of redundant information storage which improve liminal or subnormal learning.

Animals↗

Effects of frequency of feedback on transfer in concept identification.

Two experiments were performed to investigate the effects of frequency of feedback on immediate- and delayed-transfer tasks in concept identification. In the first experiment, subjects were required to learn new concepts from the same category for the transfer tasks, whereas in the second experiment the transfer tasks involved concepts from a different category. Among the major findings was that although lowering the percentage of feedback trials slows concept attainment, it facilitated transfer on all transfer tasks. In general, the fewer the number of feedback trials subjects received, the greater the amount of transfer. The results are consistent with other studies that suggest that conditions that make it more difficult for subjects to initially learn a task may have positive benefits for transfer.

Adult↗

Transfer of spatial behavior between different environments: implications for theories of spatial learning and for the role of the hippocampus in spatial learning.

In 3 experiments, rats were required to find a submerged platform located in 1 corner of an arena that had 2 long and 2 short sides; they were then trained to find the platform in a new arena that also had 2 long and 2 short sides but a different overall shape. The platform in the new arena was easier to find if it was in a corner that was geometrically equivalent, rather than the mirror image, of the corner where it had previously been located. The final experiment revealed that hippocampal lesions impaired rats' ability to find the platform in these arenas. The results suggest that rats did not use the overall shape of the arena to locate the platform but relied on more local cues and that the hippocampus plays a role in navigation based on these cues.

Animals↗

Holding on: parents' perceptions of premature infants' transfers.

OBJECTIVE: To describe parents' perceptions of their infants' transfers within a regional system of perinatal care. DESIGN: Qualitative; grounded theory methodology. SETTING: Interviews were conducted in an office adjacent to a neonatal intensive care unit (NICU), in an intermediate care unit (IMCU), and in the homes of parents. PARTICIPANTS: Fifteen parents of premature infants were recruited and interviewed during the 3 days before their infants were transferred from a NICU to an IMCU or home. The 15 parents were interviewed again during the 5 days after the initial interview to learn their feelings about the transfer. DATA COLLECTION: After consent was obtained, unstructured interviews were recorded and transcribed. Analyses of the data were ongoing, and the second interviews with parents were more focused. RESULTS: Parents identified four phases that described their transfer experience. Within the four phases, four categories were identified to depict parents experiences further. The core category of holding on reflected the belief that transfer home would become a reality and normal family life eventually would ensue. CONCLUSIONS: Nurses in NICUs and IMCUs have a responsibility to educate the parents of infants at high risk. Nurses sensitive to parental perceptions of neonatal transfer can better facilitate a positive transfer experience.

Adaptation, Psychological↗

Monocular acquisition and interocular transfer in albino guinea pigs as mediated by ipsi- or contralateral fiber systems.

Monocular acquisition and interocular transfer of a horizontal-vertical discrimination were measured in 12 male albino guinea pigs with unilateral lesions of the striate cortex, and in 6 sham-operated controls. In accordance with electrophysiological and anatomical findings, the animals learned readily using the contralateral fibers, while the ipsilateral fibers were found to be incapable of mediating either acquisition, or retention, of the discrimination. The control animals learned readily, but showed incomplete transfer.

Animals↗