جستجوی مقالات مرتبط با کلیدواژه
learning algorithm
در نشریات گروه علوم پایه
تکرار جستجوی کلیدواژه learning algorithm در مقالات مجلات علمی
-
The hybrid fuzzy differential equations have a wide range of applications in science and engineering. We consider the problem of nding their numerical solutions by using a novel hybrid method based on fuzzy neural network. Here neural network is considered as a part of large eld called neural computing or soft computing. The proposed algorithm is illustrated by numerical examples and the results obtained using the scheme presented here agree well with the analytical solutions.Keywords: Hybrid fuzzy differential equations, Fuzzy neural networks, Learning algorithm
-
In this paper, we interpret a two-point initial value problem for a second order fuzzy differential equation. We investigate a problem of finding a numerical approximation of the solution by using fuzzy neural network. Here neural network is considered as a part of a larger field called neural computing or soft computing. Finally, we illustrate our approach on an applied example in engineering.
Keywords: Second order fuzzy differential equation, fuzzy neural networks, learning algorithm -
In this paper, a novel hybrid method based on learning algorithm of fuzzy neural network and Newton-Cotes methods with positive coefficient for the solution of linear Fredholm integro-differential equation of the second kind with fuzzy initial value is presented. Here neural network is considered as a part of large field called neural computing or soft computing. We propose a learning algorithm from the cost function for adjusting fuzzyweights. This paper is one of the first attempts to derive learning algorithms from fuzzy neural networks with real input, fuzzy output, and fuzzy weights. Finally, we illustrate our approach by numerical examples.Keywords: Fuzzy neural networks, Fuzzy linear Fredholm integro, differential, Feedforward neural network, Learning algorithm
-
This paper intends to offer a new iterative method based on arti cial neural networks for finding solution of a fuzzy equations system. Our proposed fuzzi ed neural network is a ve-layer feedback neural network that corresponding connection weights to output layer are fuzzy numbers. This architecture of arti cial neural networks, can get a real input vector and calculates its corresponding fuzzy output. In order to nd the approximate solution of the fuzzy system that supposedly has a real solution, rst a cost function is de ned for the level sets of the fuzzy network and target output. Then a learning algorithm based on the gradient descent method is used to adjust the crisp input signals. The present method is illustrated by several examples with computer simulations.Keywords: System of fuzzy equations, Fuzzy feed, back neural network (FFNN), Cost function, Learning algorithm
-
In this paper, we interpret a fuzzy differential equation by using the strongly generalized differentiability concept. Utilizing the Generalized characterization Theorem. Then a novel hybrid method based on learning algorithm of fuzzy neural network for the solution of differential equation with fuzzy initial value is presented. Here neural network is considered as a part of large eld called neural computing or soft computing. The model nds the approximated solution of fuzzy differential equation inside of its domain for the close enough neighborhood of the fuzzy initial point. We propose a learning algorithm from the cost function for adjusting of fuzzy weights.Keywords: Fuzzy neural networks, Fuzzy di erential equations, Feedforward neural network, Learning algorithm
-
نشریه علوم زمین، پیاپی 81 (پاییز 1390)، صص 31 -36در این پژوهش، برآورد الگوریتم های یادگیری مختلف در شبکه عصبی برای برآورد عیار در سامانه مس پورفیری سوناجیل مقایسه شده است. هدف این پژوهش، بهینه کردن ساختار شبکه مورد استفاده و ارائه روند بهینه سازی ساختاری آن برای برآورد عیار مس برای شناسایی بهتر منطقه است. بر این اساس، دوازده الگوریتم یادگیری پس انتشار خطا برای این هدف بررسی شدند. نتایج مطالعه بیانگر آن است که در الگوریتم های مورد استفاده دو الگوریتم LM و BFG بهترین کارایی را دارند. دلایل برای نشان دادن کارایی تقریبا مساوی الگوریتم های یادگیری دیگر به صورت کمی بیان شده است. متغیرهای ورودی شبکه، موقعیت طول و عرض جغرافیایی و خروجی آن، عیار کانسار در آن مختصات است. همچنین برای به دست آوردن ساختار بهینه شبکه مورد نظر از شبکه های با تعداد لایه های مختلف استفاده شد که در پایان شبکه با تعداد دوازده نرون مورد استفاده قرار گرفت. برای بررسی تاثیر شکل عادی کردن داده ها از شکل های مختلف داده ها استفاده شد که داده های عادی شده در بازه]1 0 [نتایج بهتری داشتند. در پایان برای بهینه تر شدن شبکه همچنین از توابع مختلف انتقال در این شبکه استفاده شد که تابع انتقال تانژانت سیگموییدی با کمترین خطای ممکن همراه بود و این تابع به عنوان تابع بهینه برگزیده شد. با در نظر گرفتن شرایط بهینه مقدار R2 برای شبکه 946/0 به دست آمد که نویدگر استفاده از شبکه های عصبی با ساختار بهینه برای بهبود شرایط برآورد است.
کلید واژگان: شبکه عصبی، ساختار بهینه، برآورد عیار، الگوی یادگیری، سوناجیلIn the present research, comparative evaluation of various learning algorithms in neural network modeling was performed for ore grade estimation in Sonjail porphyry copper deposit. The main goal of the following investigation would be optimizing the network architecture and to present an architectural optimization trend to better performing the copper grade estimation within the region. Therefore, 12 algorithms were investigated back propagation learning algorithms. Based on this research it is merged that by applying the LM and BFG algorithms, there would be the best performance. The reasons why the other algorithms have the same performance would be presented within the paper as well. The input parameters are coordinates and the outputs are the copper grades for each specified point. To obtain the optimal structure, a network with different layers has been applied, which it has acquired 12 neurons within one layer. To investigate the data normal shapes, various normal shape has been acquired in the [0 1], which could merged the best results. Finally to get the best network optimizations several transfer functions have been applied, and the sigmoid transfer function illustrated least error when the transfer function is selected. Considering the optimal conditions, the R2 value has merged 0.946 for network which could be the result showing that the optimal network architecture causes estimation improvement.Keywords: Neural network, Optimal architecture, Grade estimation, Learning algorithm, Sonajil
نکته
- نتایج بر اساس تاریخ انتشار مرتب شدهاند.
- کلیدواژه مورد نظر شما تنها در فیلد کلیدواژگان مقالات جستجو شدهاست. به منظور حذف نتایج غیر مرتبط، جستجو تنها در مقالات مجلاتی انجام شده که با مجله ماخذ هم موضوع هستند.
- در صورتی که میخواهید جستجو را در همه موضوعات و با شرایط دیگر تکرار کنید به صفحه جستجوی پیشرفته مجلات مراجعه کنید.