Open Access Open Access  Restricted Access Access granted  Restricted Access Subscription Access

Vol 25, No 4 (2016)

Article

Fast online algorithm for nonlinear support vector machines and other alike models

Kecman V.

Abstract

Paper presents a unique novel online learning algorithm for eight popular nonlinear (i.e., kernel), classifiers based on a classic stochastic gradient descent in primal domain. In particular, the online learning algorithm is derived for following classifiers: L1 and L2 support vector machines with both a quadratic regularizer wtw and the l1 regularizer |w|1; regularized huberized hinge loss; regularized kernel logistic regression; regularized exponential loss with l1 regularizer |w|1 and Least squares support vector machines. The online learning algorithm is aimed primarily for designing classifiers for large datasets. The novel learning model is accurate, fast and extremely simple (i.e., comprised of few coding lines only). Comparisons of performances of the proposed algorithm with the state of the art support vector machine algorithm on few real datasets are shown.

Optical Memory and Neural Networks. 2016;25(4):203-218
pages 203-218 views

Oscillatory neural associative memories with synapses based on memristor bridges

Tarkov M.S.

Abstract

An approach to the implementation of electronic associative memories with tunable weights based on the resistor bridges containing memristors—a bidirectional associative memory (BAM) and an associative memory based on the Hopfield network—is proposed. These memories we implement as a networks of coupled phase oscillators. The conditions for the use of the operational amplifier in a comparator mode for implementing the step activation function are determined. It is shown how to use the CMOS transistors switches to control the memristance value. The experiments using LTSPICE models show that for the reference binary images with size 3 × 3 the proposed networks converges to the reference images (and, accordingly, to their inversion) with a random uniform distribution of binary pixel values of the input images. In all experiments we have no error states in spite of the number of reference patterns exceeds the classical estimations for traditional BAM and Hopfield networks.

Optical Memory and Neural Networks. 2016;25(4):219-227
pages 219-227 views

Multilevel neural net adaptive models using the metagraph approach

Fedorenko Y.S., Gapanyuk Y.E.

Abstract

The paper considers adaptive models enabling real-time processing of data flows. The drawbacks of current algorithms are examined. A method that combines advantages of deep learning, self-organizing neural nets and the metagraph approach is offered for designing adaptive models. A part of the method is realized, data clustering experiments are carried out and experimental results are analyzed.

Optical Memory and Neural Networks. 2016;25(4):228-235
pages 228-235 views

Artificial neural network approach for LNA design of GPS receiver

Singh S., Chopra P.K.

Abstract

Paper presents an ANN modeling of microwave LNA for the global positioning front end receiver, operating at 1.57542 GHz. To design LNA, multilayer perceptron architecture is used. The scattering parameters of LNA are calculated using Levenberg Marquardt Backpropagation Algorithm for the frequency range 100 MHz to 8 GHz. The inputs given to this architecture are drain to source current, drain to source voltage, temperature and frequency and the outputs are maximum available gain, noise figure and scattering parameters (magnitude as well as angle). ANN model is trained using Agilent MGA 72543 GaAs pHEMT Low Noise Amplifier datasheet and this model shows high regression. The smith and polar charts are plotted for frequency range 100 MHz to 8 GHz.

Optical Memory and Neural Networks. 2016;25(4):236-242
pages 236-242 views

Dynamic model of information processing and/or self-excitation in thalamo-cortical neuron-like models

Nuidel I.V., Sokolov M.E., Yakhno V.G.

Abstract

A functional model of signal transformation in neural networks of similar interconnected active elements is developed. The model describes dynamics of processing of sensory signals in thalamo- cortical neural networks. Using the model normal modes of signal processing are obtained (by the example of images). In addition, we reproduced regimes of propagation of areas of an increased impulse activity (analogue of epilepsy). Control of a coupling coefficient between the model’s modules reproduces a phenomenon of intermittency in thalamo-cortical oscillatory EEG patterns.

Optical Memory and Neural Networks. 2016;25(4):243-254
pages 243-254 views

Application of optoelectronic micro-displays for holographic binary data recorder based on computer generated fourier holograms

Odinokov S.B., Zlokazov E.Y., Betin A.Y., Donchenko S.S., Starikov R.S., Verenikina N.M.

Abstract

Present article highlights the researches provided during the development of holographic memory system based on application of computational methods to encode binary data pages as amplitude computer generated Fourier holograms (CGFH). Using electro-optical micro-display and projection optics CGFH can be recorded onto photosensitive medium. The type of display that is used in data recorder determines the specificities of optical scheme architecture with its own limitations and advantages. There are three projection schemes of binary data recorder discussed in the paper. A linear scheme based on transparency-type liquid crystal spatial light modulator (LC SLM), a scheme based on reflection-type liquid crystal on silicon spatial light modulator (LCoS SLM) which uses a beamsplitting cube, and the most compact and simple scheme based on self-emitting OLED-display. The results of experimental implementation of all the three projection schemes for CGFH of binary data pages record onto the holographic carrier and consequent optical reconstruction and analysis of the encoded data are presented.

Optical Memory and Neural Networks. 2016;25(4):255-261
pages 255-261 views

Holographic memory without reference beam

Shoydin S.A.

Abstract

The paper summarises the results of innovations in holographic memories without formed reference beam. The author suggests a modular system of units to read/write information to volume holograms without reference beams. The results of experimental data recording and retrieval are shown for single holograms. The author also determines expected capacities and maximum performance of the said devices.

Optical Memory and Neural Networks. 2016;25(4):262-267
pages 262-267 views

Multilayered glass composites light transmission studies

Atkarskaya A.B., Privalov V.E., Shemanin V.G.

Abstract

It has been established the total light transmission dependence on the number and structure of the layers in a film and also such a composite light transmission at the edges of the visible spectral range T1 (short wavelength—violet) and T2 (long wavelength—red) dependences from this studies. The increasing of the layers number leads to reduction of the composites light transmission. It has been reached the minimal transmission about of 2% for violet spectrum range at 24 layers in composition with this alternation of SiO2 and TiO2 layers, and the titan dioxide layer was finished this composition with the average total transmission about 10%.

Optical Memory and Neural Networks. 2016;25(4):268-271
pages 268-271 views

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies