The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The state monitoring issue of the induced draft fan in a thermal power plant by employing the gravitational searching algorithm optimized BP neural network is investigated in this paper. A new method to estimate the air quantity of the induced draft air of a thermal power plant is proposed based on the historical operation data extracted from the supervisory information system (SIS). In order to predict...
Storage reliability of the ammunition dominates the efforts in achieving the mission reliability goal. Prediction of storage reliability is important in practice to monitor the ammunition quality. In this paper we provided an integrated method where particle swarm optimization (PSO) algorithm is applied to adjust and optimize the BP neural network global parameters (weights and thresholds). The experiment...
Food safety is one of the hot issues in all over the world. It is related to national economy and people's livelihood. In recent years, food safety accidents occur in China frequently, so an effective food safety network public opinion early warning model is necessary and imperative. Therefore, the model of Back Propagation neural network based on Analytic Hierarchy Process (AHP-BP) is proposed. The...
Dropout is a very effective way of regularizing neural networks. Stochastically “dropping out” units with a certain probability discourages over-specific co-adaptations of feature detectors, preventing overfitting and improving network generalization. Besides, Dropout can be interpreted as an approximate model aggregation technique, where an exponential number of smaller networks are averaged in order...
Training deep neural networks is difficult for the pathological curvature problem. Re-parameterization is an effective way to relieve the problem by learning the curvature approximately or constraining the solutions of weights with good properties for optimization. This paper proposes to reparameterize the input weight of each neuron in deep neural networks by normalizing it with zero-mean and unit-norm,...
The inverse kinematics solution of the Six-DOF Serial Robots is a highly complex nonlinear problem, and the existence of the solution is not unique, therefore, the inverse kinematics method has attracted extensive attention. In this paper, BP neural network is applied to the inverse kinematics of Six-DOF Serial Robots, and the process of inverse solution is transformed into training of network weights...
In this paper, an approach to evaluating game states of a collectible card game Hearthstone is described. A deep neural network is employed to predict the probability of winning associated with a given game state. Encoding the game state as an input vector is based on another neural network, an autoencoder with a sparsity-inducing loss. The autoencoder encodes minion information in a sparse-like fashion...
This paper presents a deep analysis of literature on the problems of optimization of parameters and structure of the neural networks and the basic disadvantages that are present in the observed algorithms and methods. As a result, there is suggested a new algorithm for neural network structure optimization, which is free of the major shortcomings of other algorithms. The paper describes a detailed...
This project explored fundamental methods to find the factors that can be used in classifying and detecting the type of wood. Whereas, the literatures have been reviewed to determine the algorithms developed. Some experiments have been conducted to analyze the model and system. The experiments are based on artificial neural network (ANN) algorithm that used back propagation and conjugate gradient...
This paper describes the use of convolutional neural network(CNN) method to classify various image and photo of Indonesia ancient temple. The method itself implements Deep Learning technique designed for Computer Vision task. The idea behind CNN is image pre-processing through a stack of convolution layers to create many patterns that can be easily recognized. The result shows that the learning model...
In a computer vision system, handwritten digits recognition is a complex task that is central to a variety of emerging applications. It has been widely used by machine learning and computer vision researchers for implementing practical applications like computerized bank check numbers reading. In this study, we implemented a multi-layer fully connected neural network with one hidden layer for handwritten...
Road traffic accident is a serious threat to human life and safety of living environment. In this paper, a new road traffic accident prediction model (TAP-CNN) is established by using traffic accident influencing factors, such as traffic flow, weather, light to build a state matrix to describe the traffic state and CNN model. This paper uses samples to test the accuracy of the new model. The experimental...
In this work, we investigate the robustness of 1-transistor-1-resistor (1T1R) synaptic array to implement a low-precision neural network. The experimental results on 1 kb HfOx-based RRAM array show a large on/off ratio (i.e. > 105×) and 5 stable resistance states can be reliably achieved with 10× window between adjacent two states. As the RRAM has the resistance drift over time under read voltage...
The paper presents a deep analysis of the literature on the problems of optimization of parameters and the structure of the neural networks and the basic disadvantages that are present in the observed algorithms and methods. As a result, there are suggested a new algorithm for neural network structure optimization, which is devoided of the major shortcomings of other algorithms. The paper includes...
The aim of this work is the detection of solar photovoltaic panels in low-quality satellite photos. It is important to receive the geospatial data (such as country, zip code, street and home number) of installed solar panels, because they are connected directly to the local power. It will be helpful to estimate a power capacity and an energy production using the satellite photos. For this purpose,...
Communication is changing to a wireless world. Any wireless communication system needs radio frontends (transceivers) to link higher layer signals to the air interface. The increasing standard and technology diversity require transceivers with high flexibility supporting many of frequencies, standards and signal requirements. Design to cost, the demand for highest energy efficiency and MIMO systems...
Different from training common neural networks (NNs) for inference on general-purpose processors, the development of NNs for neuromorphic chips is usually faced with a number of hardware-specific restrictions. This paper proposes a systematic methodology to address the challenge. It can transform an existing trained, unrestricted NN (usually for software execution substrate) into an equivalent network...
A human action recognition method is introduced that detects a set of actions in videos by a temporal expansion with hidden Markov models of a pose detection with an artificial neural network. The method was set-up and tested using eleven actions from the MOCAP motion capture database comprising 3,947 frames. A poses alphabet of fourteen relevant poses was defined to be learned by an artificial neural...
In this paper, we propose a Wavelet Neural Network (WNN) classifier for breast cancer. WNN is a new kind of artificial neural network which is coming more popular these days. This method is based on the Wavelet Transform (WT) and classical neural networks. This paper explains how WNN classifies and uses formulas. The results of the experiments made to obtain the best results and the parameters affecting...
This paper provides a voice transformation model that uses pitch data and Feed-forward Neural Networks on Line Spectral Frequency. The aim of this work is to achieve the transformation of a speech signal produced by a source speaker by modifying voice individuality parameters such that it appears to be spoken by a chosen target speaker, without modifying the message contents. Most of the previous...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.