scholarly journals APPLICATION OF MULTIDIMENSIONAL TIME MODEL TO DYNAMICAL RELATION OF POISSON SPIKE TRAINS IN NEURAL ION CURRENT MODELS AND FORMATION OF NON-CANONICAL BASES, ISLANDS, AND G-QUADRUPLEXES IN DNA, MRNA, AND RNA AT OR NEAR THE TRANSCRIPTION

2021 ◽  
Vol 9 (1) ◽  
pp. 7-15
Author(s):  
Michael Fundator
Keyword(s):  
Neural Networks ◽  
Cpg Islands ◽  
Spike Trains ◽  
Stochastic Methods ◽  
Hölder Continuous ◽  
Time Model ◽  
Canonical Bases ◽  
Rigorous Approach ◽  
G Quadruplex ◽  
Multidimensional Methods

Ground breaking application of mathematics and biochemistry to explain formation of non-canonical bases, islands, G-quadruplex structures, and analog bases in DNA and mRNA at or near the transcription with connection to neural networks is implemented using statistical and stochastic methods apparatus with the addition of quantum principles. As a result the usual transience of Poisson spike trains (PST) becomes very instrumental tool for finding periodical type of solutions to Fokker-Plank (FP) stochastic differential equation (SDE). The present study develops new multidimensional methods of finding solutions to SDE. This is based on more rigorous approach to mathematical apparatus through Kolmogorov-Chentsov continuity theorem (KCCT) that allows the stochastic processes with jumps under certain conditions to have  γ-Holder continuous modification, which is used as basis for finding analogous parallels in dynamics of formation of CpG and non-CpG islands (CpGI or non CpGI), repeats of G-quadruplexes, and non canonical bases during DNA (de)- methylation and neural networks.

scholarly journals Biologically Plausible Sequence Learning with Spiking Neural Networks

2020 ◽  
Vol 34 (02) ◽  
pp. 1316-1323
Author(s):  
Zuozhu Liu ◽  
Thiparat Chotibut ◽  
Christopher Hillar ◽  
Shaowei Lin
Keyword(s):  
Neural Networks ◽  
Sequence Learning ◽  
Nervous Activity ◽  
Learning Rule ◽  
Generative Models ◽  
Spike Timing ◽  
Hopfield Network ◽  
Spiking Neural Networks ◽  
Theoretical Ground ◽  
Time Model

Motivated by the celebrated discrete-time model of nervous activity outlined by McCulloch and Pitts in 1943, we propose a novel continuous-time model, the McCulloch-Pitts network (MPN), for sequence learning in spiking neural networks. Our model has a local learning rule, such that the synaptic weight updates depend only on the information directly accessible by the synapse. By exploiting asymmetry in the connections between binary neurons, we show that MPN can be trained to robustly memorize multiple spatiotemporal patterns of binary vectors, generalizing the ability of the symmetric Hopfield network to memorize static spatial patterns. In addition, we demonstrate that the model can efficiently learn sequences of binary pictures as well as generative models for experimental neural spike-train data. Our learning rule is consistent with spike-timing-dependent plasticity (STDP), thus providing a theoretical ground for the systematic design of biologically inspired networks with large and robust long-range sequence storage capacity.

scholarly journals A Review of Binarized Neural Networks

Electronics ◽  
2019 ◽  
Vol 8 (6) ◽  
pp. 661 ◽  
Author(s):  
Taylor Simons ◽  
Dah-Jye Lee
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Execution Time ◽  
Deep Neural Networks ◽  
Time Model ◽  
Bitwise Operations

In this work, we review Binarized Neural Networks (BNNs). BNNs are deep neural networks that use binary values for activations and weights, instead of full precision values. With binary values, BNNs can execute computations using bitwise operations, which reduces execution time. Model sizes of BNNs are much smaller than their full precision counterparts. While the accuracy of a BNN model is generally less than full precision models, BNNs have been closing accuracy gap and are becoming more accurate on larger datasets like ImageNet. BNNs are also good candidates for deep learning implementations on FPGAs and ASICs due to their bitwise efficiency. We give a tutorial of the general BNN methodology and review various contributions, implementations and applications of BNNs.

scholarly journals Supervised Learning Algorithm for Multilayer Spiking Neural Networks with Long-Term Memory Spike Response Model

2021 ◽  
Vol 2021 ◽  
pp. 1-16
Author(s):  
Xianghong Lin ◽  
Mengwei Zhang ◽  
Xiangwen Wang
Keyword(s):  
Neural Networks ◽  
Supervised Learning ◽  
Learning Algorithm ◽  
Spike Trains ◽  
Spiking Neural Networks ◽  
Long Term Memory ◽  
Response Model ◽  
Spike Response ◽  
Spike Response Model

As a new brain-inspired computational model of artificial neural networks, spiking neural networks transmit and process information via precisely timed spike trains. Constructing efficient learning methods is a significant research field in spiking neural networks. In this paper, we present a supervised learning algorithm for multilayer feedforward spiking neural networks; all neurons can fire multiple spikes in all layers. The feedforward network consists of spiking neurons governed by biologically plausible long-term memory spike response model, in which the effect of earlier spikes on the refractoriness is not neglected to incorporate adaptation effects. The gradient descent method is employed to derive synaptic weight updating rule for learning spike trains. The proposed algorithm is tested and verified on spatiotemporal pattern learning problems, including a set of spike train learning tasks and nonlinear pattern classification problems on four UCI datasets. Simulation results indicate that the proposed algorithm can improve learning accuracy in comparison with other supervised learning algorithms.

scholarly journals About combining determinated and stochastic approaches for prediction of the heating balance of the building for water sports

2020 ◽  
Vol 22 (1) ◽  
pp. 103-112
Author(s):  
S. V. Guzhov
Keyword(s):  
Neural Networks ◽  
Regression Analysis ◽  
Artificial Neural Networks ◽  
Initial Data ◽  
Thermal Energy ◽  
Heat Balance ◽  
Stochastic Method ◽  
Stochastic Methods ◽  
Artificial Neural ◽  
Water Sports

Forecasting the demand for thermal energy by energy complexes of buildings and structures is an urgent task. To achieve the necessary accuracy of the calculation, it is customary to use various deterministic methods based on the available changing and slightly changing data about the object of study. At the same time, statistical data can also be used in analysis by stochastic methods. The purpose of this article is to analyze the question of the admissibility of combining deterministic and stochastic approaches in order to increase the accuracy of the calculation. Formulas for calculating the components of the expenditure part of the heat balance are shown on the example of a building for water sports. Based on the above formulas, a calculation with a monthly discretization in the period from January 2009 is carried out. until January 2019. An example is given of calculating the accuracy of the forecast of demand for thermal energy through multivariate regression analysis and the use of artificial neural networks. Based on the same data, an artificial neural network was trained on seven different factors: six independent and seventh — the idealized value of the building’s heat loss through the building envelope. An example of the analysis of a building for practicing water sports shows the inadmissibility of the described approach if the same initial data are used in the deterministic and stochastic method. Results: the accuracy of the forecast made using regression analysis increases with an increase in the number of factors. However, the use of an additional group of factors in the stochastic method, for example, which are numerically processed climate data that are already used as initial data, will lead to an unreasonable overestimation of the significance of the twice used factor. The presence in the predictive models using artificial neural networks of collinearity and multicollinearity of variables does not negatively affect the forecast. Conclusion: the combination of the deterministic and stochastic approaches in preparing the predicted heat balance by using only the same input data that is used in the stochastic approach in the deterministic approach is unacceptable.

scholarly journals Solution of Ruin Probability for Continuous Time Model Based on Block Trigonometric Exponential Neural Network

Symmetry ◽  
2020 ◽  
Vol 12 (6) ◽  
pp. 876
Author(s):  
Yinghao Chen ◽  
Chun Yi ◽  
Xiaoliang Xie ◽  
Muzhou Hou ◽  
Yangjin Cheng
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Ruin Probability ◽  
Risk Model ◽  
Insurance Company ◽  
Classical Risk Model ◽  
Continuous Time Model ◽  
Time Model ◽  
Trial Solution ◽  
Premium Income

The ruin probability is used to determine the overall operating risk of an insurance company. Modeling risks through the characteristics of the historical data of an insurance business, such as premium income, dividends and reinvestments, can usually produce an integral differential equation that is satisfied by the ruin probability. However, the distribution function of the claim inter-arrival times is more complicated, which makes it difficult to find an analytical solution of the ruin probability. Therefore, based on the principles of artificial intelligence and machine learning, we propose a novel numerical method for solving the ruin probability equation. The initial asset u is used as the input vector and the ruin probability as the only output. A trigonometric exponential function is proposed as the projection mapping in the hidden layer, then a block trigonometric exponential neural network (BTENN) model with a symmetrical structure is established. Trial solution is set to meet the initial value condition, simultaneously, connection weights are optimized by solving a linear system using the extreme learning machine (ELM) algorithm. Three numerical experiments were carried out by Python. The results show that the BTENN model can obtain the approximate solution of the ruin probability under the classical risk model and the Erlang(2) risk model at any time point. Comparing with existing methods such as Legendre neural networks (LNN) and trigonometric neural networks (TNN), the proposed BTENN model has a higher stability and lower deviation, which proves that it is feasible and superior to use a BTENN model to estimate the ruin probability.

Prévision hydrologique par réseaux de neurones artificiels : état de l'art

1999 ◽  
Vol 26 (3) ◽  
pp. 293-304 ◽  
Author(s):  
Paulin Coulibaly ◽  
François Anctil ◽  
Bernard Bobée
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Stochastic Methods ◽  
Multilayer Perceptrons ◽  
Neural Network Approach ◽  
Hydrological Forecasting ◽  
Learning Rules ◽  
The Neural Network ◽  
Artificial Neural ◽  
Réseaux De Neurones Artificiels

Artificial neural networks (ANN) are a novel approximation method for complex systems especially useful when the well-known statistical methods are not efficient. The multilayer perceptrons have been mainly used for hydrological forecasting over the last years. However, the connectionist theory and language are not much known to the hydrologist communauty. This paper aims to make up this gap. The ANN architectures and learning rules are presented to allow the best choice of their application. Stochastic methods and the neural network approach are compared in terms of methodology steps in the context of hydrological forecasting. Recent applications in hydrology are documented and discussed in the conclusion.Key words: artificial neural networks, hydrological forecasting, stochastic models, multilayer perceptrons.

Real-time model of three-dimensional dynamic reattachment using neural networks

1995 ◽  
Vol 32 (6) ◽  
pp. 1177-1182 ◽  
Author(s):  
William E. Faller ◽  
Scott J. Schreck ◽  
Hank E. Helin
Keyword(s):  
Neural Networks ◽  
Real Time ◽  
Three Dimensional ◽  
Time Model

ADAPTIVE SYNCHRONIZATION FOR UNKNOWN STOCHASTIC CHAOTIC NEURAL NETWORKS WITH MIXED TIME-DELAYS BY OUTPUT COUPLING

2008 ◽  
Vol 22 (24) ◽  
pp. 2391-2409 ◽  
Author(s):  
YANG TANG ◽  
JIAN-AN FANG ◽  
SUOJUN LU ◽  
QINGYING MIAO
Keyword(s):  
Neural Networks ◽  
Time Delays ◽  
Distributed Delay ◽  
Adaptive Synchronization ◽  
Output Coupling ◽  
Stochastic Neural Networks ◽  
Unknown Parameters ◽  
Chaotic Neural Networks ◽  
Rigorous Approach ◽  
Mixed Time Delays

This paper is concerned with the synchronization problem for a class of stochastic neural networks with unknown parameters and mixed time-delays via output coupling. The mixed time-delays comprise the time-varying delay and distributed delay, and the neural networks are subjected to stochastic disturbances described in terms of a Brownian motion. Firstly, we use Lyapunov functions to establish general theoretical conditions for designing the output coupling matrix. Secondly, by using the adaptive feedback technique, a simple, analytical and rigorous approach is proposed to synchronize the stochastic neural networks with unknown parameters and mixed time-delays. Finally, numerical simulation results are given to show the effectiveness of the proposed method.

scholarly journals General-Purpose Computation with Neural Networks: A Survey of Complexity Theoretic Results

2003 ◽  
Vol 15 (12) ◽  
pp. 2727-2778 ◽  
Author(s):  
Jiří Šíma ◽  
Pekka Orponen
Keyword(s):  
Neural Networks ◽  
Relevant Literature ◽  
Network Models ◽  
Network Size ◽  
General Purpose ◽  
Neural Network Models ◽  
Time Model ◽  
Continuous State ◽  
Computational Aspects ◽  
Winner Take All

We survey and summarize the literature on the computational aspects of neural network models by presenting a detailed taxonomy of the various models according to their complexity theoretic characteristics. The criteria of classification include the architecture of the network (feedforward versus recurrent), time model (discrete versus continuous), state type (binary versus analog), weight constraints (symmetric versus asymmetric), network size (finite nets versus infinite families), and computation type (deterministic versus probabilistic), among others. The underlying results concerning the computational power and complexity issues of perceptron, radial basis function, winner-take-all, and spiking neural networks are briefly surveyed, with pointers to the relevant literature. In our survey, we focus mainly on the digital computation whose inputs and outputs are binary in nature, although their values are quite often encoded as analog neuron states. We omit the important learning issues.

Sign in / Sign up

Export Citation Format

Share Document