Search PubMed⌕ Search

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.

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

At least 127 records · Page 7Linked to original sources
PubMedDate not supplied

Spontaneous recovery.

Spontaneous recovery from extinction is one of the most basic phenomena of Pavlovian conditioning. Although it can be studied by using a variety of designs, some procedures are better than others for identifying the involvement of underlying learning processes. A wide range of different learning mechanisms has been suggested as being engaged by extinction, most of which have implications for the nature of spontaneous recovery. However, despite the centrality of the notion of spontaneous recovery to the understanding of extinction, the empirical literature on its determinants is relatively sparse and quite mixed. Its very ubiquity suggests that spontaneous recovery has multiple sources.

Animals↗
PubMed2004

Gaussian processes for machine learning.

Gaussian processes (GPs) are natural generalisations of multivariate Gaussian random variables to infinite (countably or continuous) index sets. GPs have been applied in a large number of fields to a diverse range of ends, and very many deep theoretical analyses of various properties are available. This paper gives an introduction to Gaussian processes on a fairly elementary level with special emphasis on characteristics relevant in machine learning. It draws explicit connections to branches such as spline smoothing models and support vector machines in which similar ideas have been investigated. Gaussian process models are routinely used to solve hard machine learning problems. They are attractive because of their flexible non-parametric nature and computational simplicity. Treated within a Bayesian framework, very powerful statistical methods can be implemented which offer valid estimates of uncertainties in our predictions and generic model selection procedures cast as nonlinear optimization problems. Their main drawback of heavy computational scaling has recently been alleviated by the introduction of generic sparse approximations.13,78,31 The mathematical literature on GPs is large and often uses deep concepts which are not required to fully understand most machine learning applications. In this tutorial paper, we aim to present characteristics of GPs relevant to machine learning and to show up precise connections to other "kernel machines" popular in the community. Our focus is on a simple presentation, but references to more detailed sources are provided.

Algorithms↗
PubMed1999

Webs, cell assemblies, and chunking in neural nets: introduction.

This introduction to Wickelgren (1992), describes a theory of idea representation and learning in the cerebral cortex and seven properties of Hebb's (1949) formulation of cell assemblies that have played a major role in all such neural net models. Ideas are represented in the cerebral cortex by webs (innate cell assemblies), using sparse coding with sparse, all-or-none, innate linking. Recruiting a web to represent a new idea is called chunking. The innate links that bind the neurons of a web are basal dendritic synapses. Learning modifies the apical dendritic synapses that associate neurons in one web to neurons in another web.

Brain↗
PubMed2001

Distribution of tyrosine hydroxylase-containing neurons and fibers in the brain of the budgerigar (Melopsittacus undulatus): general patterns and labeling in vocal control nuclei.

The distribution of tyrosine hydroxylase (TH) was mapped out in cells and fibers of the budgerigar (Melopsittacus undulatus) brain. Special attention was given to vocal control and auditory nuclei because budgerigars are a psittacine species in which both males and females are capable of lifelong vocal learning (Farabaugh et al. [1994] J. Comp. Psychol 108:81-92). The results show that TH staining in the central nucleus of the anterior archistriatum (AAc) resembled that of surrounding archistriatal fields, except for portions of the ventral archistriatum, which exhibited substantially more TH+ fibers. Fewer fibers and fiber baskets are present in the central nucleus of the lateral neostriatum (NLc) than in surrounding fields. Both the oval nuclei of the ventral hyperstriatum (HVo) and anterior neostriatum (NAo) exhibit less fiber staining than surrounding fields whereas fiber staining in the medial NAo (NAom) and magnicellular nucleus of the parolfactory lobe (LPOm) resemble that of surrounding fields. Staining in primary telencephalic auditory nuclei was extremely low. The only sex difference observed was slightly increased TH staining in LPOm of females compared with surrounding fields on some tissue sections. These findings are in contrast to previous findings in zebra finch (Poephila guttata), a close ended vocal learning songbird in which TH staining in vocal nuclei increases during development and remains greater than surrounding fields throughout adulthood. The present results therefore support the view that catecholamines act to inhibit vocal plasticity in adult vocal learning species. Several unique features of TH-immunoreactive (ir) cell groups were observed in the brainstem including sparsely scattered TH-ir somata immediately adjacent to the third ventricle, within the tectum, basal forebrain, archistriatum, and caudal neostriatum, and in the hippocampus. These latter populations have not been described in other avian species and resemble features of the catecholamine system generally found in either reptiles or mammals.

Animals↗
PubMed1994

Immunocytochemical localization of nitric oxide synthase in the brain of the chicken.

The distribution of neuronal nitric oxide synthase (NOS) in the chicken brain was investigated by immunocytochemistry. Strongly stained neurones were concentrated in the paleostriatum augmentatum, lobus parolfactorius, ventral pallidum, olfactory tubercle, parts of the neostriatum, mesencephalic reticular formation and locus coeruleus. Cells in these areas have previously been shown to stain for NADPH-diaphorase, a histochemical activity associated with NOS. However, various structures which NADPH-diaphorase staining has suggested to contain NOS were not immunoreactive: these included the glomeruli of the olfactory bulb, magnocellular preoptic neurones, the median eminence, subcommissural organ and mesencephalic trigeminal neurones. NOS was sparsely present in the hyperstriatum ventrale, providing evidence against the involvement of nitric oxide in certain forms of learning and memory processes known to occur in this region.

Amino Acid Oxidoreductases↗
PubMed1987

Rarefaction and resolution, stimulus area and detection: an order of priority.

The aim of this research was to establish the priority of different effects between phenomenal variables in acuity tasks. The phenomenon we considered--observed at first by Vicario in 1971--is a good example of links between different percepts. In Exp. 1 the threshold of the resolution distance of square gratings having the same phenomenal frequencies (as observed in a 1985 experiment by Vardabasso and Zanuttini), although different areas, was checked. The distance at which the oblique lines, with the same phenomenal frequencies, are distinguished does not differ significantly for the two squares. In Exp. 2 we checked the effect of two different variables--the physical spacing between the lines and the area of the pattern--on detection thresholds on a task which allowed comparing the distance at which a gap in the diagonal of the squared gratings "small and large", "sparse and dense" was detected.

Adult↗
PubMed2003

Learning higher-order structures in natural images.

The theoretical principles that underlie the representation and computation of higher-order structure in natural images are poorly understood. Recently, there has been considerable interest in using information theoretic techniques, such as independent component analysis, to derive representations for natural images that are optimal in the sense of coding efficiency. Although these approaches have been successful in explaining properties of neural representations in the early visual pathway and visual cortex, because they are based on a linear model, the types of image structure that can be represented are very limited. Here, we present a hierarchical probabilistic model for learning higher-order statistical regularities in natural images. This non-linear model learns an efficient code that describes variations in the underlying probabilistic density. When applied to natural images the algorithm yields coarse-coded, sparse-distributed representations of abstract image properties such as object location, scale and texture. This model offers a novel description of higher-order image structure and could provide theoretical insight into the response properties and computational functions of lower level cortical visual areas.

Learning↗
PubMed2002

Compositional syntax from cultural transmission.

A growing body of work demonstrates that syntactic structure can evolve in populations of genetically identical agents. Traditional explanations for the emergence of syntactic structure employ an argument based on genetic evolution: Syntactic structure is specified by an innate language acquisition device (LAD). Knowledge of language is complex, yet the data available to the language learner are sparse. This incongruous situation, termed the "poverty of the stimulus," is accounted for by placing much of the specification of language in the LAD. The assumption is that the characteristic structure of language is somehow coded genetically. The effect of language evolution on the cultural substrate, in the absence of genetic change, is not addressed by this explanation. We show that the poverty of the stimulus introduces a pressure for compositional language structure when we consider language evolution resulting from iterated observational learning. We use a mathematical model to map the space of parameters that result in compositional syntax. Our hypothesis is that compositional syntax cannot be explained by understanding the LAD alone: Compositionality is an emergent property of the dynamics resulting from sparse language exposure.

Cultural Evolution↗
PubMed2005

A learning rule for the emergence of stable dynamics and timing in recurrent networks.

Neural dynamics within recurrent cortical networks is an important component of neural processing. However, the learning rules that allow networks composed of hundreds or thousands of recurrently connected neurons to develop stable dynamical states are poorly understood. Here I use a neural network model to examine the emergence of stable dynamical states within recurrent networks. I describe a learning rule that can account both for the development of stable dynamics and guide networks to states that have been observed experimentally, specifically, states that instantiate a sparse code for time. Across trials, each neuron fires during a specific time window; by connecting the neurons to a hypothetical set of output units, it is possible to generate arbitrary spatial-temporal output patterns. Intertrial jitter of the spike time of a given neuron increases as a direct function of the delay at which it fires. These results establish a learning rule by which cortical networks can potentially process temporal information in a self-organizing manner, in the absence of specialized timing mechanisms.

Action Potentials↗
PubMed1998

Development of localized oriented receptive fields by learning a translation-invariant code for natural images.

Neurons in the mammalian primary visual cortex are known to possess spatially localized, oriented receptive fields. It has previously been suggested that these distinctive properties may reflect an efficient image encoding strategy based on maximizing the sparseness of the distribution of output neuronal activities or alternately, extracting the independent components of natural image ensembles. Here, we show that a strategy for transformation-invariant coding of images based on a first-order Taylor series expansion of an image also causes localized, oriented receptive fields to be learned from natural image inputs. These receptive fields, which approximate localized first-order differential operators at various orientations, allow a pair of cooperating neural networks, one estimating object identity ('what') and the other estimating object transformations ('where'), to simultaneously recognize an object and estimate its pose by jointly maximizing the a posteriori probability of generating the observed visual data. We provide experimental results demonstrating the ability of such networks to factor retinal stimuli into object-centred features and object-invariant transformation estimates.

Animals↗
PubMed2006

H-CORE: enabling genome-scale Bayesian analysis of biological systems without prior knowledge.

The Bayesian network is a popular tool for describing relationships between data entities by representing probabilistic (in)dependencies with a directed acyclic graph (DAG) structure. Relationships have been inferred between biological entities using the Bayesian network model with high-throughput data from biological systems in diverse fields. However, the scalability of those approaches is seriously restricted because of the huge search space for finding an optimal DAG structure in the process of Bayesian network learning. For this reason, most previous approaches limit the number of target entities or use additional knowledge to restrict the search space. In this paper, we use the hierarchical clustering and order restriction (H-CORE) method for the learning of large Bayesian networks by clustering entities and restricting edge directions between those clusters, with the aim of overcoming the scalability problem and thus making it possible to perform genome-scale Bayesian network analysis without additional biological knowledge. We use simulations to show that H-CORE is much faster than the widely used sparse candidate method, whilst being of comparable quality. We have also applied H-CORE to retrieving gene-to-gene relationships in a biological system (The 'Rosetta compendium'). By evaluating learned information through literature mining, we demonstrate that H-CORE enables the genome-scale Bayesian analysis of biological systems without any prior knowledge.

Algorithms↗
PubMed1998

Chills and thrills: does radio harm our children? The controversy over program violence during the age of radio.

A review of the popular and scientific periodical for the 1930s and 1940s revealed that the controversy surrounding the radio "thriller" and its possible harmful effects on young listeners was one of radio's most highly publicized issues during its golden years of broadcasting. Many of the questions raised concerning this issue were similar to those asked later during the age of television. Relying heavily upon the psychoanalytic emphasis on emotion, catharsis, and intrapsychic dynamics, expert opinion voiced in various popular periodicals and newspapers of the day suggested that the violence and excitement portrayed in many of the crime and adventure programs was harmless, and perhaps beneficial, for most listeners. However, research in support of this conclusion was sparse, and psychologists evidenced little interest in the issue. Not until the advent of television, and the emergence of social learning theory in the early 1960s, did psychologists direct significant research effort towards evaluating the effects of media violence.

Child↗
PubMed2004

Synaptic transformations underlying highly selective auditory representations of learned birdsong.

Stimulus-specific neuronal responses are a striking characteristic of several sensory systems, although the synaptic mechanisms underlying their generation are not well understood. The songbird nucleus HVC (used here as a proper name) contains projection neurons (PNs) that fire temporally sparse bursts of action potentials to playback of the bird's own song (BOS) but are essentially silent when presented with other acoustical stimuli. To understand how such remarkable stimulus specificity emerges, it is necessary to compare the auditory-evoked responsiveness of the afferents of HVC with synaptic responses in identified HVC neurons. We found that inactivating the interfacial nucleus of the nidopallium (NIf) could eliminate all auditory-evoked subthreshold activity in both HVC PN types, consistent with NIf serving as the major auditory afferent of HVC. Simultaneous multiunit extracellular recordings in NIf and intracellular recordings in HVC revealed that NIf population activity and HVC subthreshold responses were similar in their selectivity for BOS and that NIf spikes preceded depolarizations in all HVC cell types. These results indicate that information about the BOS as well as other auditory stimuli is transmitted synaptically from NIf to HVC. Unlike HVC PNs, however, HVC-projecting NIf neurons fire throughout playback of BOS as well as non-BOS stimuli. Therefore, temporally sparse BOS-evoked firing and enhanced BOS selectivity, manifested as an absence of suprathreshold responsiveness to non-BOS stimuli, emerge in HVC. The transformation to a sparse auditory representation parallels differences in NIf and HVC activity patterns seen during singing, which may point to a common mechanism for encoding sensory and motor representations of song.

Animals↗
PubMed2000

Auditory responses in the vocal motor system of budgerigars.

Budgerigars (Melopsittacus undulatus) are small Australian parrots that can imitate novel sounds in adulthood and therefore serve as a convenient model system for the study of vocal learning in adult animals. Previous anatomical studies had indicated that known auditory regions in the telencephalon of budgerigars are connected, albeit indirectly and rather sparsely, to vocal motor nuclei. Physiological evidence for connections between the auditory and vocal motor systems in budgerigars had been lacking, however. Here, we show that neurons in a telencephalic vocal motor region, i.e., the central nucleus of the lateral neostriatum (NLc), are responsive to auditory stimuli in isoflurane-anesthetized budgerigars. These responses are highly variable from trial to trial and frequently have latencies in excess of 100 ms. Neurons in NLc generally respond better to a budgerigar's own contact call than to a white noise stimulus, but the response preferences of NLc neurons in budgerigars are generally weaker and more diverse than the response preferences of neurons in the high vocal center of songbirds, which is probably analogous to NLc. These data indicate that parrots and songbirds, which have evolved the ability to learn vocalizations independently of one another, have both evolved physiologically effective connections between their auditory and vocal motor systems. Interestingly, however, the anatomical pathways by which the auditory and vocal motor systems interact, and the physiological details of how they communicate, appear to be significantly different between the two taxa.

Acoustic Stimulation↗
PubMed2006

Unsupervised analysis of polyphonic music by sparse coding.

We investigate a data-driven approach to the analysis and transcription of polyphonic music, using a probabilistic model which is able to find sparse linear decompositions of a sequence of short-term Fourier spectra. The resulting system represents each input spectrum as a weighted sum of a small number of "atomic" spectra chosen from a larger dictionary; this dictionary is, in turn, learned from the data in such a way as to represent the given training set in an (information theoretically) efficient way. When exposed to examples of polyphonic music, most of the dictionary elements take on the spectral characteristics of individual notes in the music, so that the sparse decomposition can be used to identify the notes in a polyphonic mixture. Our approach differs from other methods of polyphonic analysis based on spectral decomposition by combining all of the following: (a) a formulation in terms of an explicitly given probabilistic model, in which the process estimating which notes are present corresponds naturally with the inference of latent variables in the model; (b) a particularly simple generative model, motivated by very general considerations about efficient coding, that makes very few assumptions about the musical origins of the signals being processed; and (c) the ability to learn a dictionary of atomic spectra (most of which converge to harmonic spectral profiles associated with specific notes) from polyphonic examples alone-no separate training on monophonic examples is required.

Journal Article↗
PubMed2006

Critical thinking as an outcome of context-based learning among post RN students: a literature review.

Increasingly complex needs and expanding roles in the delivery of health care require professional nurses to be capable critical thinkers and self-directed learners. Problem-based learning (PBL) is promoted as a mean to facilitate critical thinking (CT) in nursing students attending generic and post RN baccalaureate programs. The authors summarize and analyze nursing research and theoretical literature (1992-2005) related to four key topic areas: critical thinking, nursing education, PBL, and post RN education, to determine what is known about the impact of PBL on CT among post RN students. CINAHL, ERIC, Medline and PsychINFO databases were searched and based on the literature retrieved, there is no strong research evidence to suggest changes occur in baccalaureate nursing students' CT during their educational programs, including those using PBL. Similar results were found among post RN students although the literature on this topic is sparse. There is a need for additional research and ongoing development of valid and reliable instruments to measure CT in nurses to guide teaching and learning strategies that effectively facilitate CT among nursing students.

Education, Nursing↗
PubMed1996

Modeling the spontaneous reactivation of experience-specific hippocampal cell assembles during sleep.

During slow-wave sleep (SWS) following periods of spatial activity, hippocampal place cells that were temporally correlated, by virtue of the overlap of their place fields, exhibit enhanced temporal correlations, even though the animal sleeps in a different location (Wilson and McNaughton [1994] Science 267:676-679). The discharge of cells with overlapped place fields is more correlated in subsequent sleep, particularly during sharp waves, than in sleep episodes prior to the behavior, or than cell pairs with non-overlapped place fields. The reactivated correlated states appear during hippocampal sharp waves (SPWs), and are weak or absent in the inter-SPW interval. A simple conceptual hypothesis for this phenomenon is developed, based on the idea that hippocampal place fields reflect a two-dimensional distribution of continuously overlapping dynamic attractors in which each location is represented by the self-sustaining activity of a small subset of neurons with overlapping place fields. A numerical simulation of this hypothesis, based on a simplified representation of the CA8 recurrent network, accounts qualitatively for the main observations, including SPW-like dynamics. It is shown that, under conditions in which the connection patterns have been previously established, either associative or nonassociative mechanisms might underlie the reactivation of recently experienced states. These two alternatives appear, under at least some conditions (e.g., sparse coding), to be indistinguishable.

Animals↗