Применение нейронных сетей прямого распространения для формирования оценок параметров по методу максимального правдоподобия
This paper introduces an approach for implementing maximum likelihood parameter estimators using feedforward artificial neural network of multilayer perceptron architecture. A theoretical foundation of the proposed approach is presented in the assumption that the model of observation is known as well as the values of its vector of parameters. For a practical example the implementation of direction of arrival estimator for the active ring antenna array is shown. In order to estimate a performance and accuracy of the proposed approach, the results of numerical calculation are presented, compared to the algorithm based on optimal numerical solution and referenced to Cramer–Rao lower bound. The results also indicate that there is no significant dependency of the accuracy of estimation on actual parameter value. Moreover, the calculations take significantly less time, although some of it is spent on the initial training of the neural network.
For the stochastic differential equationdX(t)=a X ( t ) + b X ( t - 1 )dt+dW(t),t≥0,
the local asymptotic properties of the likelihood function are studied. They depend strongly on the true value of the parameter ϑ=(a,b) * . Eleven different cases are possible if ϑ runs through ℝ 2 . Let ϑ ^ T be the maximum likelihood estimator of ϑ based on (X(t), t≥T). Applications to the asymptotic behaviour of ϑ ^ T as T→∞ are given.
The paper is devoted to the description of a new multi-purpose intellectual decision support system. We present the algorithms used and the results achieved in applying the system to analyzing and forecasting the sea ice area in the Northern Hemisphere. The impact of solar radiation on the changes in the sea ice area was confirmed. Application of interval neural nets to medium-term forecasting of sea ice area changes was justified.
This paper deals with the angle-of-arrival passive location system consisting of mutually coherent circular antenna arrays implementing one-step maximum likelihood based estimation procedure. The question raised in the present paper is accuracy degradation due to the arbitrary positions of spectral components in the locator receiver bandwidth in comparison with Cramer-Rao Lower Bound for the case of Gaussian random radio emission. The well-known expression of angle estimator accuracy in the case of band-uniform spectrum is compared to the results obtained by the numerical statistical modeling; this allows one to evaluate the deterioration of the accuracy due to arbitrary positioned spectral components. The numerical modeling also provides the discussion for preliminary time-window filtration of the received signals in order to increase the accuracy of angle estimator. Thus it was shown that the common used windows such as Hann, Bartlett and Keiser-Bessel can be useful in the case of sparse arbitrary spectrum.
Describes the history of establishment of the scientific field of "artificial intelligence", describes the main directions of its development and scope of mapping the three main strategic approaches to creation of intellectual systems: the technology of expert systems, neural network technology and technologies of evolutionary modelling. Theoretical bases and examples of the development of intelligent systems and examples of their application to the data mining, in Economics, in business, in psychology, in sociology and other fields. The book builds and expands the worldview of the graduates, improving his ranking in life. At the same time, the book is a comprehensive guide to mastering technologies for creating neural network intelligent systems and their applications for solving a wide range of problems encountered in many fields of human activity. For students of higher educational institutions.
We strengthen the convergence result in our paper, ibid. 5, No. 6, 1059-1098 (1999; Zbl 0983.62049), proving the local asymptotic mixed normality property in one of the 11 cases considered in that paper.
This proceedings bring together contributions from researchers from academia and industry to report the latest cutting edge research made in the areas of Fuzzy Computing, Neuro Computing and hybrid Neuro-Fuzzy Computing in the paradigm of Soft Computing. The FANCCO 2015 conference explored new application areas, design novel hybrid algorithms for solving different real world application problems. After a rigorous review of the 68 submissions from all over the world, the referees panel selected 27 papers to be presented at the Conference. The accepted papers have a good, balanced mix of theory and applications. The techniques ranged from fuzzy neural networks, decision trees, spiking neural networks, self organizing feature map, support vector regression, adaptive neuro fuzzy inference system, extreme learning machine, fuzzy multi criteria decision making, machine learning, web usage mining, Takagi-Sugeno Inference system, extended Kalman filter, Goedel type logic, fuzzy formal concept analysis, biclustering etc. The applications ranged from social network analysis, twitter sentiment analysis, cross domain sentiment analysis, information security, education sector, e-learning, information management, climate studies, rainfall prediction, brain studies, bioinformatics, structural engineering, sewage water quality, movement of aerial vehicles, etc.
ICCV is the premier international computer vision event comprising the main conference and several co-located workshops and tutorials. With its high quality and low cost, it provides an exceptional value for students, academics and industry researchers.
Generalized error-locating codes are discussed. An algorithm for calculation of the upper bound of the probability of erroneous decoding for known code parameters and the input error probability is given. Based on this algorithm, an algorithm for selection of the code parameters for a specified design and input and output error probabilities is constructed. The lower bound of the probability of erroneous decoding is given. Examples of the dependence of the probability of erroneous decoding on the input error probability are given and the behavior of the obtained curves is explained.
This proceedings publication is a compilation of selected contributions from the “Third International Conference on the Dynamics of Information Systems” which took place at the University of Florida, Gainesville, February 16–18, 2011. The purpose of this conference was to bring together scientists and engineers from industry, government, and academia in order to exchange new discoveries and results in a broad range of topics relevant to the theory and practice of dynamics of information systems. Dynamics of Information Systems: Mathematical Foundation presents state-of-the art research and is intended for graduate students and researchers interested in some of the most recent discoveries in information theory and dynamical systems. Scientists in other disciplines may also benefit from the applications of new developments to their own area of study.
Let G be a semisimple algebraic group whose decomposition into the product of simple components does not contain simple groups of type A, and P⊆G be a parabolic subgroup. Extending the results of Popov , we enumerate all triples (G, P, n) such that (a) there exists an open G-orbit on the multiple flag variety G/P × G/P × . . . × G/P (n factors), (b) the number of G-orbits on the multiple flag variety is finite.
I give the explicit formula for the (set-theoretical) system of Resultants of m+1 homogeneous polynomials in n+1 variables