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 199 records · Page 11Linked to original sources
PubMed2002

Learning to report.

RATIONALE AND OBJECTIVES: In radiology residency programs it is the task of residents to learn how to effectively report the findings and impressions from patient images they review, generally from an apprenticeship design. The authors sought to learn residents' perceptions of the value this apprenticeship model holds for them. MATERIALS AND METHODS: Thirty-eight residents were interviewed in a 1-month field study and asked four questions about their experiences: How did you learn to report? What formal or organized dictation instruction have you received? What feedback do you receive? What is your opinion about your experience? RESULTS: It was found that residents perceive that the apprenticeship model is inadequate when learning to report. The authors found reporting issues could be classified into three categories: perceived lack of transferable learning from observation, lack of explicit direction from faculty, and need for a more structured approach to learning. CONCLUSION: Although residents agreed that learning the skills to report would be gained eventually through the apprenticeship model, they also believed that it did not give them the confidence or competence a more structured program could provide.

Clinical Competence↗
PubMed2026

Predictive design of tissue-specific mammalian enhancers that function in the mouse embryo.

Enhancers control tissue-specific gene expression across animals1. Although deep learning2,3 has enabled enhancer prediction and design in mammalian cell lines and non-mammalian model organisms4-10 (reviewed in a previous publication11), it remains unclear whether such approaches can operate within the regulatory complexity of mammalian genomes and tissues in vivo. Here we present a general strategy for designing tissue-specific enhancers that function reliably in mice. We use deep learning to train compact convolutional neural networks on curated chromatin accessibility data and fine-tune them by transfer learning on validated human and mouse enhancers. Guided by these models, we design 15 synthetic enhancers for the heart, limb and central nervous system in mouse embryos, all of which are active in their intended target tissue. These results demonstrate that mammalian enhancer function can be reliably inferred from DNA sequence alone, enabling the predictive de novo design of tissue-specific synthetic enhancers from modest training sets. This work establishes a generalizable framework for programmable control of mammalian gene expression in vivo, opening new avenues in functional genomics, synthetic biology and gene therapy.

Animals↗
PubMed2006

On No-KR tests in motor learning, retention and transfer.

In this paper we examine the role of no-knowledge of results (KR) tests in motor learning, retention and transfer. A distinction is drawn between studies of KR and studies of other issues in motor learning that use a No-KR test. Arguments are presented against the currently received position that a No-KR test is an essential protocol to examine motor learning, although a No-KR test is most relevant in the few contexts where the outcome information is not naturally available. Perspectives are provided on the limited though proper role of No-KR tests in motor learning, retention and transfer and these are illustrated through experimental examples. It is proposed that the current restrictive interpretation of KR as only augmented outcome information is narrowing not only the context of KR research but also the study of motor learning in general.

Conditioning, Classical↗
PubMed2006

Discrimination learning induced by training with identical stimuli.

Sensory stimuli become easier to detect or distinguish with practice. It is generally assumed that the task-relevant stimulus dimension becomes increasingly more salient as a result of attentively performing the task at a level that is neither too easy nor too difficult. However, here we show improved auditory frequency discrimination following training with physically identical tones that were impossible to discriminate. We also show that learning transfers across tone frequencies and across modalities: training on a silent visuospatial computer game improved thresholds on the auditory discrimination task. We suggest that three processes are necessary for optimal perceptual learning: sensitization through exposure to the stimulus, modality- and dimension-specific attention, and general arousal.

Acoustic Stimulation↗
PubMed1999

Implicit motor sequence learning is not purely perceptual.

Many reports have indicated that implicit learning of sequences in a choice response time task is primarily perceptual; subjects learn the sequence of stimuli rather than the sequence of motor responses. Three experiments tested whether implicit motor sequence learning could be purely perceptual: no support was found for that hypothesis. Subjects who merely watched stimuli did not learn the sequence implicitly (Experiment 1), and sequence learning transferred robustly to a different set of stimulus cues (Experiment 2). In the final experiment, the stimulus-response mapping was changed at transfer so that one group of subjects pushed the same sequence of keys but saw new stimuli, whereas another group pushed a different sequence of keys but saw the same stimuli. Transfer to the new mapping was shown only if the motor sequence was kept constant, not the perceptual sequence. It is proposed that subjects learn a sequence of response locations in this and similar tasks.

Adult↗
PubMed1998

Simultaneous learning of motion discrimination in two directions.

We take issue with theories about the direction specificity in perceptual learning of motion discrimination. Trials of motion discrimination in two opposite directions were interleaved in uneven proportions (2:1). Human subjects improved faster in the direction with less frequent trials, indicating that learning transferred from the more frequent to the less frequent direction.

Adult↗
PubMedDate not supplied

Recovery of function after brain damage: the extrinsic stimulus specificity of spared neural traces.

Previous research has demonstrated that while primary visual cortex lesions in rats will disrupt performance of a learned brightness discrimination, the neural traces associated with this discrimination are spared. Moreover, the spared neural traces may, in certain instances, influence the postoperative reinstatement of a brightness discrimination. The present research was designed to determine the generality of this postoperative influence. The results indicate that while the normal rat and the visual decorticate rat may readily transfer learning between discriminations involving different discriminative cues, a visual lesion between two discriminations prevents this transfer. These data are interpreted as indicating that the postoperative influence of a spared neural trace is dependent upon the reoccurrence of specific extrinsic stimulus cues.

Animals↗
PubMed1992

Left hemisphere preponderance in trajectorial learning.

The present study examines the organisational principle of a trajectorial storage mechanism for isochronically controlled overlearned movements. 48 right-handed subjects were trained on new ideograms at five different sizes. After the trajectory had been overtrained with one hand, learning transfer to the contralateral hand was analysed. Results indicate a significant transfer of trajectorial information after training with the right hand towards the left hand which points to a preponderance of the dominant left hemisphere in generalizing storage of learned trajectories to either upper extremity. The trajectorial storage system has clear geometrical restraints.

Female↗
PubMed1985

State-dependent learning in hyperactive children receiving methylphenidate.

State-dependent learning refers to a failure of learning mastered under one drug condition to be remembered when tested under another drug condition. Previous studies of state-dependent learning in hyperactive children receiving stimulants have yielded conflicting results. The authors systematically evaluated learning and transfer of learning in children who were or were not receiving methylphenidate and included several design features intended to optimize the likelihood of demonstrating state-dependent learning. They found no evidence of state-dependent learning. These results diminish concern regarding state-dependent effects in hyperactive children who are positive drug responders and who are clinically administered methylphenidate to control their hyperactivity.

Attention Deficit Disorder with Hyperactivity↗
PubMed1980

Teaching about transference: a videotape introduction.

A seminar utilizing a videotaped course of actual consecutive psychotherapy sessions with an experienced therapist and his patient has unique advantages in teaching concepts of transference. The experience provides a role model that demonstrates the therapeutic use of transference. Learning takes place by identification with the therapist on cognitive, affective, and experiential levels.

Internship and Residency↗
PubMed2004

How learning one category influences the learning of another: intercategory generalization based on analogy and specific stimulus information.

We investigated the effect of learning one category structure on the learning of a related category structure. Photograph-name combinations, called identifiers, were associated with values of four demographic attributes. Two problems were related by analogous demographic attributes, common identifiers, or both to examine the impact of common identifier, related general characteristics, and the interaction of the two variables in mediating learning transfer from one category structure to another. Problems sharing the same identifier information prompted greater positive transfer than those not sharing the same identifier information. In contrast, analogous defining characteristics in the two problems did not facilitate transfer. We computed correlations between responses to first-problem stimuli and responses to analogous second-problem stimuli for each participant. The analogous characteristics produced a tendency to respond in the same way to corresponding stimuli in the two problems. The results support an alignment between category structures related by analogous defining characteristics, which is facilitated by specific identifier information shared by two category structures.

Generalization, Psychological↗
PubMed1991

Hemispheric asymmetries in visual pattern processing in infancy.

A right hemisphere advantage was observed in a previous study of 4- to 9-month old infants presented with a face discrimination task (de Schonen & Mathivet, 1990). The present study was designed to investigate pattern processing by the two hemispheres and the interhemispheric communication of this processing. Infants aged 4 to 9 months were tested with divided visual field presentations in one or two discrimination tasks. Under both task conditions, the infants had to discriminate between two patterns in which only two local components differed. Under one condition the components of the patterns were arranged so as to produce a face-like pattern. Under the other condition the same components were arranged into arbitrary patterns that were not "good form" patterns. No performance asymmetry was observed with the arbitrary patterns; whereas, a right hemisphere (RH) disadvantage was observed with the face-like patterns compared with both the RH performances on the arbitary patterns and the left hemisphere (LH) performances on the face-like patterns. These results show that the RH advantage for individual face recognition is not due to a general immaturity or inability of the LH in pattern processing at this period of development, nor to a more specific inability in a local mode of pattern processing. On the other hand, the RH does not completely lack local processing capacity, but is at a disadvantage when this local mode of processing has to be used with face-like (or good form) patterns. The interhemispheric communication of visual discrimination learning was tested by measuring learning transfer between the visual fields. Contrary to de Schonen and Bry's study (1987) on faceness recognition, no data in favor of interhemispheric communication were recorded in the present study.

Attention↗
PubMed2000

Perceptual learning in visual search generalizes over tasks, locations, and eyes.

In a visual search task, targets containing elementary features are detected in parallel, while a serial search is necessary for the detection of a target without a feature, or for targets containing conjunctions of features. In this study, we re-investigated the role of practice in visual search tasks, using an uncued visual search paradigm. Under some circumstances, initially serial tasks can become parallel with practice. Perceptual learning of feature search tasks is rapid (a few hundreds of trials are sufficient to transform serial into parallel search), long-lasting (a learned task is retained over several months), but far less specific than learning of other visual tasks (see also Sireteanu & Rettenbach, 1995a [Vision Research, 35, 2037-2043]). Learning transfers from one task to another, from one location in the visual field to another, and between the two eyes of a given subject, even if the subject has reduced stereopsis. Search for a conjunction of orientation and colour becomes more efficient, suggesting that a different search strategy emerges after prolonged practice. These results suggest that learning of visual search tasks modifies neural structures located at a high level in the visual pathway, involving different, presumably more central neural circuits, than the learning of visual discriminations and hyperacuity.

Analysis of Variance↗
PubMed2026

A novel deep learning-driven framework for improving lncRNA comprehensive annotation with LncADeep 2.0.

MOTIVATION: Long non-coding RNAs (lncRNAs) have emerged as crucial players in diverse physiological and pathological processes, yet the biological mechanisms of the vast majority of lncRNAs remain elusive. To fill this gap, it is necessary to improve the accuracy of lncRNA identification and functional annotation. RESULTS: Here, we introduce LncADeep 2.0, an integrated deep learning framework designed to meet these needs. In the identification module, LncADeep 2.0 incorporated novel peptide features along with sequence and structural information, demonstrating superior performance over our previous LncADeep and other existing tools on both annotated transcripts from GENCODE and RNA-seq data. For functional annotation, LncADeep 2.0 leveraged lncRNA-centric interaction networks and gene ontology terms through the transfer learning strategy to achieve robust annotation performance with limited functional data. Compared to LncADeep, LncADeep 2.0 could accurately elucidate the general functions of given lncRNA sequences, predict tissue- or cell-type-specific functions from bulk and single-cell RNA-seq data, and establish connections between tumor-associated lncRNAs and genomic markers. Overall, LncADeep 2.0 stands out as an efficient and reliable tool for lncRNA identification and functional annotation across a wide spectrum of biological processes. AVAILABILITY AND IMPLEMENTATION: LncADeep 2.0 is available for use at https://github.com/Jefferson-Chou/LncADeep2 and https://doi.org/10.5281/zenodo.17164767.

RNA, Long Noncoding↗
PubMed2025

OpenSpliceAI: An efficient, modular implementation of SpliceAI enabling easy retraining on non-human species.

The SpliceAI deep learning system is currently one of the most accurate methods for identifying splicing signals directly from DNA sequences. However, its utility is limited by its reliance on older software frameworks and human-centric training data. Here we introduce OpenSpliceAI, a trainable, open-source version of SpliceAI implemented in PyTorch to address these challenges. OpenSpliceAI supports both training from scratch and transfer learning, enabling seamless retraining on species-specific datasets and mitigating human-centric biases. Our experiments show that it achieves faster processing speeds and lower memory usage than the original SpliceAI code, allowing large-scale analyses of extensive genomic regions on a single GPU. Additionally, OpenSpliceAI's flexible architecture makes for easier integration with established machine learning ecosystems, simplifying the development of custom splicing models for different species and applications. We demonstrate that OpenSpliceAI's output is highly concordant with SpliceAI. In silico mutagenesis (ISM) analyses confirm that both models rely on similar sequence features, and calibration experiments demonstrate similar score probability estimates.

Journal Article↗
PubMed1979

Clinical therapy for accommodative responses: transfer effects upon performance.

A clinical therapy program featuring accommodative training was administered to a group of children with diagnosed disorders of accommodative function. The children ranged in age from six to twelve years. A group of subjects representing the same clinical population, and not differing significantly in age or grade level, acted as a control group. The control subjects participated in a therapy program of a similar duration, wherein perceptual-motor training (unrelated to the training of accommodative skills) was administered. A nearpoint pencil-and-paper task was administered to all subjects before and after the training programs, to assess changes in performance as a criterion of learning transfer and behavioral generalization. A significantly greater decrease in errors occurred in the group receiving the accommodative training as contrasted to the control group. No significant differences were found in the time scores. The results suggest that accommodative training, for children with diagnosed accommodative disorders, has transfer effects upon nearpoint performance relating to improved accuracy.

Accommodation, Ocular↗