Learning real-world stimuli in a neural network with spike-driven synaptic dynamics

We present a model of spike-driven synaptic plasticity inspired by experimental observations and motivated by the desire to build an electronic hardware device that can learn to classify complex stimuli in a semisupervised fashion. During training, patterns of activity are sequentially imposed on the input neurons, and an additional instructor signal drives the output neurons toward the desired activity. The network is made of integrate-and-fire neurons with constant leak and a floor. The synapses are bistable, and they are modified by the arrival of presynaptic spikes. The sign of the change is determined by both the depolarization and the state of a variable that integrates the postsynaptic action potentials. Following the training phase, the instructor signal is removed, and the output neurons are driven purely by the activity of the input neurons weighted by the plastic synapses. In the absence of stimulation, the synapses preserve their internal state indefinitely. Memories are also very robust to the disruptive action of spontaneous activity. A network of 2000 input neurons is shown to be able to classify correctly a large number (thousands) of highly overlapping patterns (300 classes of preprocessed Latex characters, 30 patterns per class, and a subset of the NIST characters data set) and to generalize with performances that are better than or comparable to those of artificial neural networks. Finally we show that the synaptic dynamics is compatible with many of the experimental observations on the induction of long-term modifications (spike-timing-dependent plasticity and its dependence on both the postsynaptic depolarization and the frequency of pre- and postsynaptic neurons).

Calcium waves driven by "sensitization" wave-fronts

OBJECTIVE: Cellular Ca(2+) waves are understood as reaction-diffusion systems sustained by Ca(2+)-induced Ca(2+) release (CICR) from Ca(2+) stores. Given the recently discovered sensitization of Ca(2+) release channels (ryanodine receptors; RyRs) of the sarcoplasmic reticulum (SR) by luminal SR Ca(2+), waves could also be driven by RyR sensitization, mediated by SR overloading via Ca(2+) pump (SERCA), acting in tandem with CICR. METHODS: Confocal imaging of the Ca(2+) indicator fluo-3 was combined with UV-flash photolysis of caged compounds and the whole-cell configuration of the patch clamp technique to carry out these experiments in isolated guinea pig ventricular cardiomyocytes. RESULTS: Upon sudden slowing of the SERCA in cardiomyocytes with a photoreleased inhibitor, waves indeed decelerated immediately. No secondary changes of Ca(2+) signaling or SR Ca(2+) content due to SERCA inhibition were observed in the short time-frame of these experiments. CONCLUSIONS: Our findings are consistent with Ca(2+) loading resulting in a zone of RyR 'sensitization' traveling within the SR, but inconsistent with CICR as the predominant mechanism driving the Ca(2+) waves. This alternative mode of RyR activation is essential to fully conceptualize cardiac arrhythmias triggered by spontaneous Ca(2+) release.

EEG reactivity and EEG activity in never-treated acute schizophrenics, measured with spectral parameters and dimensional complexity

Our approaches to the use of EEG studies for the understanding of the pathogenesis of schizophrenic symptoms are presented. The basic assumptions of a heuristic and multifactorial model of the psychobiological brain mechanisms underlying the organization of normal behavior is described and used in order to formulate and test hypotheses about the pathogenesis of schizophrenic behavior using EEG measures. Results from our studies on EEG activity and EEG reactivity (= EEG components of a memory-driven, adaptive, non-unitary orienting response) as analyzed with spectral parameters and "chaotic" dimensionality (correlation dimension) are summarized. Both analysis procedures showed a deviant brain functional organization in never-treated first-episode schizophrenia which, within the framework of the model, suggests as common denominator for the pathogenesis of the symptoms a deviation of working memory, the nature of which is functional and not structural.

Experimental results on gravity driven fully condensing flows in vertical tubes, their agreement with theory, and their differences with shear driven flows' boundary-condition sensitivities

This doctoral thesis presents the experimental results along with a suitable synthesis with computational/theoretical results towards development of a reliable heat transfer correlation for a specific annular condensation flow regime inside a vertical tube. For fully condensing flows of pure vapor (FC-72) inside a vertical cylindrical tube of 6.6 mm diameter and 0.7 m length, the experimental measurements are shown to yield values of average heat transfer co-efficient, and approximate length of full condensation. The experimental conditions cover: mass flux G over a range of 2.9 kg/m2-s ≤ G ≤ 87.7 kg/m2-s, temperature difference ∆T (saturation temperature at the inlet pressure minus the mean condensing surface temperature) of 5 ºC to 45 ºC, and cases for which the length of full condensation xFC is in the range of 0 < xFC < 0.7 m. The range of flow conditions over which there is good agreement (within 15%) with the theory and its modeling assumptions has been identified. Additionally, the ranges of flow conditions for which there are significant discrepancies (between 15 -30% and greater than 30%) with theory have also been identified. The paper also refers to a brief set of key experimental results with regard to sensitivity of the flow to time-varying or quasi-steady (i.e. steady in the mean) impositions of pressure at both the inlet and the outlet. The experimental results support the updated theoretical/computational results that gravity dominated condensing flows do not allow such elliptic impositions.

Comparison of three methods for driven pile capacity

A significant cost for foundations is the design and installation of piles when they are required due to poor ground conditions. Not only is it important that piles be designed properly, but also that the installation equipment and total cost be evaluated. To assist in the evaluation of piles a number of methods have been developed. In this research three of these methods were investigated, which were developed by the Federal Highway Administration, the US Corps of Engineers and the American Petroleum Institute (API). The results from these methods were entered into the program GRLWEAPTM to assess the pile drivability and to provide a standard base for comparing the three methods. An additional element of this research was to develop EXCEL spreadsheets to implement these three methods. Currently the Army Corps and API methods do not have publicly available software and must be performed manually, which requires that data is taken off of figures and tables, which can introduce error in the prediction of pile capacities. Following development of the EXCEL spreadsheet, they were validated with both manual calculations and existing data sets to ensure that the data output is correct. To evaluate the three pile capacity methods data was utilized from four project sites from North America. The data included site geotechnical data along with field determined pile capacities. In order to achieve a standard comparison of the data, the pile capacities and geotechnical data from the three methods were entered into GRLWEAPTM. The sites consisted of both cohesive and cohesionless soils; where one site was primarily cohesive, one was primarily cohesionless, and the other two consisted of inter-bedded cohesive and cohesionless soils. Based on this limited set of data the results indicated that the US Corps of Engineers method more closely compared with the field test data, followed by the API method to a lesser degree. The DRIVEN program compared favorably in cohesive soils, but over predicted in cohesionless material.