In the primary visual cortex of monkeys, the development of ocular dominance and orientation selectivity is at least partially driven by neural activity. We propose a modiied Hebb-type learning mechanism, which takes into account non-speciic components of activity-dependent synaptic modiication. It is shown analytically, that ocular dominance and ON-OFF-segregation occur simultaneously in a lin...

We assess the accuracy of various state-of-the-art methods for reconstructing gene and protein regulatory networks in the context of circadian regulation. Gene expression and protein concentration time series are simulated from a recently published regulatory network of the circadian clock in A. thaliana, which is mathematically described by a Markov jump process based on Michaelis-Menten kinet...

Recent work has shown that recurrent neural networks (RNNs) can implicitly capture and exploit hierarchical information when trained to solve common natural language processing tasks such as language modeling (Linzen et al., 2016) and neural machine translation (Shi et al., 2016). In contrast, the ability to model structured data with non-recurrent neural networks has received little attention ...

Abstract Introduction: The extremely low accident rate for U.S air carriers relative to that of general aviation (?1 and ?60/million flight hours respectively) partly reflects advanced airman certification, more demanding recurrency training stringent operational regulations. However, whether such skillset/training/regulations translate into improved safety airline pilots operating in the envir...

An artificial neural network has been trained by the error backpropagation technique to recognise phonemes and words. The speech material was recorded by a male Swedish talker and was labelled by a phonetician. There were 38 output nodes corresponding to Swedish phonemes. The training algorithm was somewhat modified to increase the training speed. Introducing coarticulation information by addin...

A criticism of neural network architectures is their susceptibility to “catastrophic interference” the ability to forget previously learned data when presented with new patterns. To avoid this, neural network architectures have been developed which specifically provide the network with a memory, either through the use of a context unit, which can store patterns for later recall, or which combin...

OBJECTIVE To construct a score that allows prediction of major depressive episode (MDE) persistence in individuals with MDE using determinants of persistence identified in previous research. METHOD Data were derived from 250 subjects from the general population with new MDE according to DSM-III-R. These subjects were recruited from the Netherlands Mental Health Survey and Incidence Study (NEM...

An artificial neural network has been trained by the error back-propagation technique to recognise phonemes and words. The speech material was recorded by a male Swedish talker and was labelled by a phonetician. There were 38 output nodes corresponding to Swedish phonemes. Introducing coarticulation information by adding simple recurrency to the net is shown to more effective than expanding the...

It is proposed that compulsive checking occurs when people who believe that they have a special, elevated responsibility for preventing harm, mainly to others, are unsure that the perceived threat has been reduced or removed. The intensity and duration of this checking is determined by three "multipliers": increased responsibility, probability of harm and anticipated seriousness of harm. The re...

Multiple sclerosis is a disease that characterized by the nervous system injury. The pathophysiology unknown, but it supposed caused an increase of immune response to myelin sheath, which causes damage in system. Actually, there are lot immunomodulatory drugs control this pathology. One most important aspects fact irreversible; therefore, use for illness just symptoms; nevertheless, some review...