Search results for "Neural"
showing 10 items of 2783 documents
A Scheme for Continuous Input to the Tsetlin Machine with Applications to Forecasting Disease Outbreaks
2019
In this paper, we apply a new promising tool for pattern classification, namely, the Tsetlin Machine (TM), to the field of disease forecasting. The TM is interpretable because it is based on manipulating expressions in propositional logic, leveraging a large team of Tsetlin Automata (TA). Apart from being interpretable, this approach is attractive due to its low computational cost and its capacity to handle noise. To attack the problem of forecasting, we introduce a preprocessing method that extends the TM so that it can handle continuous input. Briefly stated, we convert continuous input into a binary representation based on thresholding. The resulting extended TM is evaluated and analyzed…
Identification of the most informative wavelengths for non-invasive melanoma diagnostics in spectral region from 450 to 950 nm
2020
In this study 300 skin lesion (including 32 skin melanomas) multispectral data cubes were analyzed. The multi-step and single step machine learning approaches were analyzed to find the wavebands that provide the most information that helps discriminate skin melanoma from other benign pigmented lesions. The multi-step machine learning approach assumed training several models but proved itself to be ineffective. The reason for that is a necessity to train a segmentation model on a very small dataset and utilization of standard machine learning classifier which have shown poor classification performance. The single-step approach is based on a deep learning neural network. We have conducted 260…
Enabling XCSF to cope with dynamic environments via an adaptive error threshold
2020
The learning classifier system XCSF is a variant of XCS employed for function approximation. Although XCSF is a promising candidate for deployment in autonomous systems, its parameter dependability imposes a significant hurdle, as a-priori parameter optimization is not feasible for complex and changing environmental conditions. One of the most important parameters is the error threshold, which can be interpreted as a target bound on the approximation error and has to be set according to the approximated function. To enable XCSF to reliably approximate functions that change during runtime, we propose the use of an error threshold, which is adapted at run-time based on the currently achieved …
Semi-Supervised Classification Method for Hyperspectral Remote Sensing Images
2004
A new approach to the classification of hyperspectral images is proposed. The main problem with supervised methods is that the learning process heavily depends on the quality of the training data set. In remote sensing, the training set is useful only for simultaneous images or for images with the same classes taken under the same conditions; and, even worse, the training set is frequently not available. On the other hand, unsupervised methods are not sensitive to the number of labelled samples since they work on the whole image. Nevertheless, relationship between clusters and classes is not ensured. In this context, we propose a combined strategy of supervised and unsupervised learning met…
The Ultimate Fate of Supercooled Liquids
2010
In recent years it has become widely accepted that a dynamical length scale {\xi}_{\alpha} plays an important role in supercooled liquids near the glass transition. We examine the implications of the interplay between the growing {\xi}_{\alpha} and the size of the crystal nucleus, {\xi}_M, which shrinks on cooling. We argue that at low temperatures where {\xi}_{\alpha} > {\xi}_M a new crystallization mechanism emerges enabling rapid development of a large scale web of sparsely connected crystallinity. Though we predict this web percolates the system at too low a temperature to be easily seen in the laboratory, there are noticeable residual effects near the glass transition that can account …
Growing length scales in a supercooled liquid close to an interface
2002
We present the results of molecular dynamics computer simulations of a simple glass former close to an interface between the liquid and the frozen amorphous phase of the same material. By investigating F_s(q,z,t), the incoherent intermediate scattering function for particles that have a distance z from the wall, we show that the relaxation dynamics of the particles close to the wall is much slower than the one for particles far away from the wall. For small z the typical relaxation time for F_s(q,z,t) increases like exp(Delta/(z-z_p)), where Delta and z_p are constants. We use the location of the crossover from this law to the bulk behavior to define a first length scale tilde{z}. A differe…
Classical and ab-initio molecular dynamic simulation of an amorphous silica surface
2001
We present the results of a classical molecular dynamic simulation as well as of an ab initio molecular dynamic simulation of an amorphous silica surface. In the case of the classical simulation we use the potential proposed by van Beest et al. (BKS) whereas the ab initio simulation is done with a Car-Parrinello method (CPMD). We find that the surfaces generated by BKS have a higher concentration of defects (e.g. concentration of two-membered rings) than those generated with CPMD. In addition also the distribution functions of the angles and of the distances are different for the short rings. Hence we conclude that whereas the BKS potential is able to reproduce correctly the surface on the …
Perineuronal Net Formation and the Critical Period for Neuronal Maturation in the Hypothalamic Arcuate Nucleus
2019
In leptin-deficient ob/ob mice, obesity and diabetes are associated with abnormal development of neurocircuits in the hypothalamic arcuate nucleus (ARC)1, a critical brain area for energy and glucose homoeostasis2,3. Because this developmental defect can be remedied by systemic leptin administration, but only if given before postnatal day 28, a critical period for leptin-dependent development of ARC neurocircuits has been proposed4. In other brain areas, critical-period closure coincides with the appearance of perineuronal nets (PNNs), extracellular matrix specializations that restrict the plasticity of neurons that they enmesh5. Here we report that in humans and rodents, subsets of neurons…
Fasting enhances the response of arcuate neuropeptide Y-glucose-inhibited neurons to decreased extracellular glucose
2009
0363-6143 (Print) Comparative Study In Vitro Journal Article Research Support, N.I.H., Extramural; Fasting increases neuropeptide Y (NPY) expression, peptide levels, and the excitability of NPY-expressing neurons in the hypothalamic arcuate (ARC) nucleus. A subpopulation of ARC-NPY neurons ( approximately 40%) are glucose-inhibited (GI)-type glucose-sensing neurons. Hence, they depolarize in response to decreased glucose. Because fasting enhances NPY neurotransmission, we propose that during fasting, GI neurons depolarize in response to smaller decreases in glucose. This increased excitation in response to glucose decreases would increase NPY-GI neuronal excitability and enhance NPY neurotr…
Hepatic CB1 receptor is required for development of diet-induced steatosis, dyslipidemia, and insulin and leptin resistance in mice
2007
Diet-induced obesity is associated with fatty liver, insulin resistance, leptin resistance, and changes in plasma lipid profile. Endocannabinoids have been implicated in the development of these associated phenotypes, because mice deficient for the cannabinoid receptor CB1 (CB1-/-) do not display these changes in association with diet-induced obesity. The target tissues that mediate these effects, however, remain unknown. We therefore investigated the relative role of hepatic versus extrahepatic CB1 receptors in the metabolic consequences of a high-fat diet, using liver-specific CB1 knockout (LCB1-/-) mice. LCB1(-/-) mice fed a high-fat diet developed a similar degree of obesity as that of …