Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “representation 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 919 records · Page 51Linked to original sources
PubMed2004

Oligo kernels for datamining on biological sequences: a case study on prokaryotic translation initiation sites.

BACKGROUND: Kernel-based learning algorithms are among the most advanced machine learning methods and have been successfully applied to a variety of sequence classification tasks within the field of bioinformatics. Conventional kernels utilized so far do not provide an easy interpretation of the learnt representations in terms of positional and compositional variability of the underlying biological signals. RESULTS: We propose a kernel-based approach to datamining on biological sequences. With our method it is possible to model and analyze positional variability of oligomers of any length in a natural way. On one hand this is achieved by mapping the sequences to an intuitive but high-dimensional feature space, well-suited for interpretation of the learnt models. On the other hand, by means of the kernel trick we can provide a general learning algorithm for that high-dimensional representation because all required statistics can be computed without performing an explicit feature space mapping of the sequences. By introducing a kernel parameter that controls the degree of position-dependency, our feature space representation can be tailored to the characteristics of the biological problem at hand. A regularized learning scheme enables application even to biological problems for which only small sets of example sequences are available. Our approach includes a visualization method for transparent representation of characteristic sequence features. Thereby importance of features can be measured in terms of discriminative strength with respect to classification of the underlying sequences. To demonstrate and validate our concept on a biochemically well-defined case, we analyze E. coli translation initiation sites in order to show that we can find biologically relevant signals. For that case, our results clearly show that the Shine-Dalgarno sequence is the most important signal upstream a start codon. The variability in position and composition we found for that signal is in accordance with previous biological knowledge. We also find evidence for signals downstream of the start codon, previously introduced as transcriptional enhancers. These signals are mainly characterized by occurrences of adenine in a region of about 4 nucleotides next to the start codon. CONCLUSIONS: We showed that the oligo kernel can provide a valuable tool for the analysis of relevant signals in biological sequences. In the case of translation initiation sites we could clearly deduce the most discriminative motifs and their positional variation from example sequences. Attractive features of our approach are its flexibility with respect to oligomer length and position conservation. By means of these two parameters oligo kernels can easily be adapted to different biological problems.

Algorithms↗
PubMed2006

Dimensions of cognition in an insect, the honeybee.

This review provides evidence for the enormous richness of insect behavior, its high flexibility, and the cross-talk between different behavioral routines. The memory structure established by multiple forms of learning represents sensory inputs and relates behaviors in such a way that representations of complex environmental conditions are formed. Navigation and communication in social hymenoptera are particularly telling examples in this respect, but it is fair to conclude that similar integrated forms of dealing with the environment will be found in other insects when they are studied more closely. In this sense, research addressing behavioral complexity and its underlying neural substrates is necessary to characterize the real potential of insect learning and memory. Usually, such an approach has been used to characterize behavioral simplicity rather than complexity. It seems therefore timely to focus on the latter by studying problem solving alongside and in addition to elemental forms of learning.

Animal Communication↗
PubMed2005

Effects of stimulus structure and target-distracter similarity on the development of visual memory representations in schizophrenia.

INTRODUCTION: In a previous study (Silverstein, Bakshi, Chapman, & Nowlis, 1998a, Cognitive Neuropsychiatry) we demonstrated that while schizophrenia patients showed similar learning curves as nonpatient controls when determining whether a configural pattern of elements has been seen before or not, they did not demonstrate any learning curve with nonconfigural stimuli (in contrast to controls). Methodological limitations of that study, however, precluded generalisability of those effects. METHODS: In the present study, therefore, different groups of schizophrenia patients (n = 18) and controls (n = 22) were administered a modified version of the familiarity judgement task used in Silverstein et al. (1998a). In the new version of the task: (1) the patterns that repeated were different for each participant; and (2) half of the nonrepeating patterns were configural and half were nonconfigural. RESULTS: These indicated stronger perceptual learning effects for the configural compared to the nonconfigural patterns, and overall better performance and more pronounced learning effects for the control group. CONCLUSIONS: The data provide further evidence for a schizophrenia-related impairment in perceptual organisation and in the ability to develop memory representations for novel stimuli.

Journal Article↗
PubMed2005

A search advantage for faces learned in motion.

Recently there has been growing interest in the role that motion might play in the perception and representation of facial identity. Most studies have considered old/new recognition as a task. However, especially for non-rigid motion, these studies have often produced contradictory results. Here, we used a delayed visual search paradigm to explore how learning is affected by non-rigid facial motion. In the current studies we trained observers on two frontal view faces, one moving non-rigidly, the other a static picture. After a delay, observers were asked to identify the targets in static search arrays containing 2, 4 or 6 faces. On a given trial target and distractor faces could be shown in one of five viewpoints, frontal, 22 degrees or 45 degrees to the left or right. We found that familiarizing observers with dynamic faces led to a constant reaction time advantage across all setsizes and viewpoints compared to static familiarization. This suggests that non-rigid motion affects identity decisions even across extended periods of time and changes in viewpoint. Furthermore, it seems as if such effects may be difficult to observe using more traditional old/new recognition tasks.

Adolescent↗
PubMed1999

Phonological recoding and orthographic learning: A direct test of the self-teaching hypothesis.

According to the self-teaching hypothesis (Share, 1995), word-specific orthographic representations are acquired primarily as a result of the self-teaching opportunities provided by the phonological recoding of novel letter strings. This hypothesis was tested by asking normal second graders to read aloud short texts containing embedded pseudoword targets. Three days later, target spellings were correctly identified more often, named more quickly, and spelled more accurately than alternate homophonic spellings. Experiment 2 examined whether this rapid orthographic learning can be attributed to mere visual exposure to target strings. It was found that viewing the target letter strings under conditions designed to minimize phonological processing significantly attenuated orthographic learning. Experiment 3 went on to show that this reduced orthographic learning was not attributable to alternative nonphonological factors (brief exposure durations or decontextualized presentation). The results of a fourth experiment suggested that the contribution of pure visual exposure to orthographic learning is marginal. It was concluded that phonological recoding is critical to the acquisition of word-specific orthographic representations as proposed by the self-teaching hypothesis.

Child↗
PubMed2005

Hippocampal lesions facilitate instrumental learning with delayed reinforcement but induce impulsive choice in rats.

BACKGROUND: Animals must frequently act to influence the world even when the reinforcing outcomes of their actions are delayed. Learning with action-outcome delays is a complex problem, and little is known of the neural mechanisms that bridge such delays. When outcomes are delayed, they may be attributed to (or associated with) the action that caused them, or mistakenly attributed to other stimuli, such as the environmental context. Consequently, animals that are poor at forming context-outcome associations might learn action-outcome associations better with delayed reinforcement than normal animals. The hippocampus contributes to the representation of environmental context, being required for aspects of contextual conditioning. We therefore hypothesized that animals with hippocampal lesions would be better than normal animals at learning to act on the basis of delayed reinforcement. We tested the ability of hippocampal-lesioned rats to learn a free-operant instrumental response using delayed reinforcement, and what is potentially a related ability -- the ability to exhibit self-controlled choice, or to sacrifice an immediate, small reward in order to obtain a delayed but larger reward. RESULTS: Rats with sham or excitotoxic hippocampal lesions acquired an instrumental response with different delays (0, 10, or 20 s) between the response and reinforcer delivery. These delays retarded learning in normal rats. Hippocampal-lesioned rats responded slightly less than sham-operated controls in the absence of delays, but they became better at learning (relative to shams) as the delays increased; delays impaired learning less in hippocampal-lesioned rats than in shams. In contrast, lesioned rats exhibited impulsive choice, preferring an immediate, small reward to a delayed, larger reward, even though they preferred the large reward when it was not delayed. CONCLUSION: These results support the view that the hippocampus hinders action-outcome learning with delayed outcomes, perhaps because it promotes the formation of context-outcome associations instead. However, although lesioned rats were better at learning with delayed reinforcement, they were worse at choosing it, suggesting that self-controlled choice and learning with delayed reinforcement tax different psychological processes.

Animals↗
PubMed1997

The emergence of perceptual category representations in young infants: a connectionist analysis.

There has been recent interest in the idea that principles governing learning in connectionist networks can form the basis for an alternative understanding of developmental processes (Elman, Bates, Karmiloff-Smith, Johnson, Parisi, & Plunkett, 1996). The present paper can be viewed as a case example of the usefulness (and limitations) of connectionist modeling for the study of infant cognition. Specifically, the paper reports on a series of connectionist models designed to analyze the factors responsible for the emergence of global-level and basic-level category representations in young infants. The models (1) simulated the formation of global-level and basic-level representations, (2) revealed a global-to-basic order of category emergence, (3) uncovered the formation of two distinct global-level representations-an initial "self-organizing" perceptual global level and a subsequently "trained" arbitrary (i.e., nonperceptual) global level, and (4) displayed a gradual transition from perceptual global-level to perceptual basic-level representation with increasing exposure to training stimuli. Hypotheses for empirical investigations of category development in infants that follow from the modeling efforts are discussed.

Child Development↗
PubMed2005

Song tutoring triggers CaMKII phosphorylation within a specialized portion of the avian basal ganglia.

In several songbird species, a specialized anterior forebrain pathway (AFP) that includes part of the avian basal ganglia has been implicated specifically in song learning. To further elucidate cellular mechanisms and circuitry involved in vocal learning, we used quantitative immunoblot analysis to determine if early song tutoring promotes within the AFP phosphorylation of calcium/calmodulin-dependent kinase II (CaMKII), a multifunctional kinase whose phosphorylation at threonine 286 is critical for many forms of neural plasticity and behavioral learning. We report that in young male zebra finches likely to have begun the process of song acquisition, brief tutoring by a familiar conspecific adult promotes a dramatic increase in levels of phosphorylated CaMKII (pCaMKII) in Area X, the striatal/pallidal component of the AFP. In contrast, pCaMKII levels in this region were not elevated if 1) the tutor did not sing, 2) the tutor sang but was visually isolated from the pupil, or 3) the tutor was an unfamiliar adult. In young males that had not previously heard any conspecific song, first exposure to a song tutor produced a more modest, but significant rise in pCaMKII levels. Young females (who do not develop song behavior) did not exhibit any effect of tutoring on pCaMKII levels in that portion of the basal ganglia that corresponds to Area X in males. These data are consistent with the hypothesis that Area X participates in encoding and/or attaching reward value to a representation of tutor song that is accessed later to guide motor learning.

Acoustic Stimulation↗
PubMed2004

Dynamics of learning and transfer of muscular and spatial relative phase in bimanual coordination: evidence for abstract directional codes.

The present study addressed whether the timing of muscle activation and the relative direction of limb movements are dissociable constraints that may affect learning and transfer of bimanual coordination patterns, either independently or in combination. Subjects were assigned to two experimental groups in which the to-be-learned muscular phasing (135 degrees ) was either practiced with 45 degrees (i.e., predominantly isodirectional) or 135 degrees (i.e., predominantly nonisodirectional) of spatial relative phase (RP) across 2 days of practice. Prior to, during, and following practice, probe tests were held in which various relative phasing patterns were administered to assess transfer of learning. Converging evidence was obtained that the relative direction of moving limbs prominently constrained transfer of learning rather than muscular relationships. Acquisition of a specific pattern resulted in spontaneous positive transfer of learning to a new coordination pattern having the same spatial RP but not to a pattern with a different spatial RP, irrespective of muscular phasing relationships. In summary, the present results suggest that learning and transfer of coordination patterns is mediated by abstract directional codes that become part of the memory representation for bimanual coordination.

Adult↗
PubMed2002

Slow feature analysis: unsupervised learning of invariances.

Invariant features of temporally varying signals are useful for analysis and classification. Slow feature analysis (SFA) is a new method for learning invariant or slowly varying features from a vectorial input signal. It is based on a nonlinear expansion of the input signal and application of principal component analysis to this expanded signal and its time derivative. It is guaranteed to find the optimal solution within a family of functions directly and can learn to extract a large number of decorrelated features, which are ordered by their degree of invariance. SFA can be applied hierarchically to process high-dimensional input signals and extract complex features. SFA is applied first to complex cell tuning properties based on simple cell output, including disparity and motion. Then more complicated input-output functions are learned by repeated application of SFA. Finally, a hierarchical network of SFA modules is presented as a simple model of the visual system. The same unstructured network can learn translation, size, rotation, contrast, or, to a lesser degree, illumination invariance for one-dimensional objects, depending on only the training stimulus. Surprisingly, only a few training objects suffice to achieve good generalization to new objects. The generated representation is suitable for object recognition. Performance degrades if the network is trained to learn multiple invariances simultaneously.

Algorithms↗
PubMed2006

Perceptual learning without feedback in non-stationary contexts: data and model.

The role of feedback in perceptual learning is probed in an orientation discrimination experiment under destabilizing non-stationary conditions, and explored in a neural-network model. Experimentally, perceptual learning was examined with periodic alteration of a strong external noise context. The speed of learning, the performance loss at each change in external noise context (switch cost), and the asymptotic accuracy d' without feedback were very similar or identical to those with feedback. However, lack of feedback led to higher decision bias (error responses matching the external noise context). In the model, the stimulus representations are constant, whereas the read-out connections to a decision unit learn by a Hebbian plasticity rule that may be augmented by additional feedback input and criterion control of decision bias.

Feedback, Psychological↗
PubMed1999

Faithful representations with topographic maps.

Topographic map algorithms that are aimed at building "faithful representations" also yield maps that transfer the maximum amount of information available about the distribution from which they receive input. The weight density (magnification factor) of these maps is proportional to the input density, or the neurons of these maps have an equal probability to be active (equiprobabilistic map). As MSE minimization is not compatible with equiprobabilistic map formation in general, a number of heuristics have been devised in order to compensate for this discrepancy in competitive learning schemes, e.g. by adding a "conscience" to the neurons' firing behavior. However, rather than minimizing a modified MSE criterion, we introduce a new unsupervised competitive learning rule, called the kernel-based Maximum Entropy learning Rule (kMER), for topographic map formation, that optimizes an information-theoretic criterion directly. To each neuron a radially symmetric kernel is associated, with a given center and radius, and the two are updated in such a way that the (unconditional) information-theoretic entropy of the neurons' outputs is maximized. We review a number of competitive learning rules for building equiprobabilistic maps. As benchmark tests for the faithfulness of the representations, we consider two types of distributions and compare the performances of these rules and kMER, for batch and incremental learning. As a first example application, we consider non-parametric density estimation where the maps are used for generating "pilot" estimates in kernel-based density estimation. The second application we envisage for kMER is "on-line" adaptive filtering of speech signals, using Gabor functions as wavelet filters. The topographic feature maps that are developed in this way differ in several respects from those obtained with Kohonen's Adaptive-Subspace SOM algorithm.

Journal Article↗
PubMed2002

Learning abstract relations from using categories.

When people learn categories, the importance of the features and relations in the category representation reflects both their diagnosticity for classification and their relevance to the use of the category. In earlier work in which the influence of category use on the representation has been shown, only cases in which the features and relations were simple, observable, and very specific were examined. Learners may begin to understand the underlying similarities of category members by using the categories. In the four experiments presented here, learners applied a simple category-specific formula to category members. The test results showed that the learners had incorporated relations among features from this use, including cases in which the relations were abstract. This learning occurred even though the relations were actually not predictive of category membership but just perceived to be so as a function of the use.

Adult↗
PubMed2005

Orbitofrontal cortex, associative learning, and expectancies.

Orbitofrontal cortex is characterized by its unique pattern of connections with subcortical areas, such as basolateral amygdala. Here we distinguish between the critical role of these areas in associative learning and the pivotal contribution of OFC to the manipulation of this information to control behavior. This contribution reflects the ability of OFC to signal the desirability of expected outcomes, which requires the integration of associative information with information concerning internal states and goals in representational memory.

Amygdala↗
PubMed2025

Whole-genome phenotype prediction with machine learning: open problems in bacterial genomics.

MOTIVATION: How can we identify causal genetic mechanisms governing bacterial traits? Initial efforts entrusting machine learning models to handle the task of predicting phenotype from genotype yield high accuracy scores. However, attempts to extract meaningful interpretations from the predictive models are found to be corrupted by falsely identified 'causal' features. Relying solely on pattern recognition and correlations is unreliable, significantly so in bacterial genomics settings where high-dimensionality and spurious associations are the norm. Though it is not yet clear whether we can overcome this hurdle, significant efforts are being made towards discovering potential high-risk bacterial genetic variants. In view of this, we set up open problems surrounding phenotype prediction from bacterial whole-genome datasets and extending those approaches to learning causal effects, and discuss challenges that impact the reliability of a machine's decision-making when faced with datasets of this nature. RESULTS: We identify major sources of non-injectivity in the formulation of the genotype-to-phenotype mapping function-linkage-disequilibrium, limited sampling, information loss in representations, unmeasured confounders and observational noise-and analyse their implications for machine learning applications. Using a collection of 4,140 Staphylococcus aureus isolates, we illustrate challenges surrounding the defined open problems. AVAILABILITY AND IMPLEMENTATION: Raw sequencing data are available from the European Nucleotide Archive (ENA) under project accessions ERP001012, PRJEB3174, PRJEB2655, PRJEB2756, and PRJEB2944. Assemblies and annotations were generated with the Sanger bacterial pipeline (https://github.com/sanger-pathogens/vr-codebase) and unitigs extracted using DBGWAS (https://gitlab.com/leoisl/dbgwas).

Machine Learning↗
PubMed2001

Abstract reward and punishment representations in the human orbitofrontal cortex.

The orbitofrontal cortex (OFC) is implicated in emotion and emotion-related learning. Using event-related functional magnetic resonance imaging (fMRI), we measured brain activation in human subjects doing an emotion-related visual reversal-learning task in which choice of the correct stimulus led to a probabilistically determined 'monetary' reward and choice of the incorrect stimulus led to a monetary loss. Distinct areas of the OFC were activated by monetary rewards and punishments. Moreover, in these areas, we found a correlation between the magnitude of the brain activation and the magnitude of the rewards and punishments received. These findings indicate that one emotional involvement of the human orbitofrontal cortex is its representation of the magnitudes of abstract rewards and punishments, such as receiving or losing money.

Adult↗
PubMed2000

Social complexity and social intelligence.

When we talk of the 'nature of intelligence', or any other attribute, we may be referring to its essential structure, or to its place in nature, particularly the function it has evolved to serve. Here I examine both, from the perspective of the evolution of intelligence in primates. Over the last 20 years, the Social (or 'Machiavellian') Intelligence Hypothesis has gained empirical support. Its core claim is that the intelligence of primates is primarily an adaptation to the special complexities of primate social life. In addition to this hypothesis about the function of intellect, a secondary claim is that the very structure of intelligence has been moulded to be 'social' in character, an idea that presents a challenge to orthodox views of intelligence as a general-purpose capacity. I shall outline the principal components of social intelligence and the environment of social complexity it engages with. This raises the question of whether domain specificity is an appropriate characterization of social intelligence and its subcomponents, like theory of mind. As a counter-argument to such specificity I consider the hypothesis that great apes exhibit a cluster of advanced cognitive abilities that rest on a shared capacity for second-order mental representation.

Animals↗
PubMed2001

Dorsal striatal head direction and hippocampal place representations during spatial navigation.

Several theories of basal ganglia function describe a striatal contribution to learning that is independent of hippocampal function. This study examined the question of whether the striatum should be regarded as functioning independently of or acting in concert with limbic structures. Dorsal striatal head direction cells and hippocampal place cells were recorded in parallel while rats performed a hippocampal-dependent radial maze task. Changes in the directional preference of head direction cells and the location of place fields were compared following alterations of the sensory environment. When familiar visual cues were presented in new spatial arrangements, or when new visual cues were placed in a familiar environment, rotations of directional preferences were consistent with the mean place-field response. When familiar visual and nonvisual cues were presented in conflict, or when rats were exposed to novel environments, the responses of the two cell types were inconsistent relative to each other. This pattern suggests that current perceptions and expectations of familiar spatial contexts may dynamically modulate the relationship between hippocampus and dorsal striatum.

Animals↗